Regulatory Enforcement Under New York's Martin Act: From

University of Chicago Law School University of Chicago Law School

Chicago Unbound Chicago Unbound

Journal Articles Faculty Scholarship

2018

Regulatory Enforcement Under New York's Martin Act: From Regulatory Enforcement Under New York's Martin Act: From

Financial Fraud to Global Warming Financial Fraud to Global Warming

Richard A. Epstein

Follow this and additional works at: https://chicagounbound.uchicago.edu/journal_articles

Part of the Law Commons

Recommended Citation Recommended Citation

Richard A. Epstein, "Regulatory Enforcement Under New York's Martin Act: From Financial Fraud to Global

Warming," 14 New York University Journal of Law and Business 805 (2018).

This Article is brought to you for free and open access by the Faculty Scholarship at Chicago Unbound. It has been

accepted for inclusion in Journal Articles by an authorized administrator of Chicago Unbound. For more

information, please contact unbound@law.uchicago.edu.

NEW YORK UNIVERSITY

JOURNAL OF LAW & BUSINESS

OLUME

14 SUMMER 2018 N

UMBER

REGULATORY ENFORCEMENT UNDER NEW

YORK’S MARTIN ACT: FROM FINANCIAL

FRAUD TO GLOBAL WARMING

ICHARD

A. E

PSTEIN

NTRODUCTION

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 806

I. T

NLEASHING OF THE

ARTIN

. . . . . . . . . . . 807

II. T

ARTIN

THE

OMMON

-L

CTION FOR

ECEIT

AND THE

EDERAL

ECURITY AND

XCHANGE

CTS

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 819

A. The Origins of the Martin Act . . . . . . . . . . . . . . . . . 819

B. The Common Law of Deceit . . . . . . . . . . . . . . . . . . . 822

C. Basis of Liability: Fraud, Recklessness, Negligence,

and Beyond . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 823

D. Statements of Fact: Materiality, Omissions,

Opinions, and Predictions . . . . . . . . . . . . . . . . . . . . . 829

1. Materiality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 830

2. Omissions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 836

3. Statements of Fact and Opinion . . . . . . . . . . . . 838

4. Predictions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 844

E. Causation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 845

F. Privity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 849

* Laurence A. Tisch Professor of Law, New York University School of

Law; Peter and Kirsten Bedford Senior Fellow, the Hoover Institution; James

Parker Hall Distinguished Service Professor of Law Emeritus and Senior Lec-

turer, the University of Chicago Law School; Visiting Scholar, the Manhattan

Institute. James Copland of the Manhattan Institute provided exhaustive

commentary on earlier drafts of this paper. My thanks to Connor Haynes,

Carmiel Schickler, Samuel Schoenburg, and Nathan Yaffe, 2016, New York

University School of Law, and Jonathan Povilonis, 2018, New York University

School of Law, for their excellent research assistance. Thanks also to the

Manhattan Institute for financial support for this project.

805

806 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

III. T

ALLEABLE

ARTIN

CT IN

CTION

. . . . . . . . . 867

A. Busting Bogus Financial Schemes . . . . . . . . . . . . . . 868

B. The AIG–Greenberg Litigation . . . . . . . . . . . . . . . . . 870

C. Global Warming: Fraud and Free Speech . . . . . . . . 882

1. The Litigation . . . . . . . . . . . . . . . . . . . . . . . . . . . 882

a. ExxonMobil . . . . . . . . . . . . . . . . . . . . . . . . 882

b. Global Warming and Public

Nuisance . . . . . . . . . . . . . . . . . . . . . . . . . . . 896

2. The Science . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 903

ONCLUSION

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 918

NTRODUCTION

In this Article, I shall examine the uses and limitations of

New York State’s Martin Act, adopted in 1921,

as a means of

combating various types of financial, business, and social mis-

conduct both large and small. The major innovation of the

Martin Act was that it stripped down the basic fraud action by

eliminating the standard elements of knowledge, reliance, and

damages. Potential liability under the Martin Act therefore is

broader than that found under the common-law tort of deceit

in both cases of civil and criminal liability. In addition, there

are ways in which liability under the Martin Act is broader than

that found under federal securities and exchange laws, passed

in 1933 and 1934,

which themselves rely in modified form on

the basic system of common-law actions.

In Part I, I shall outline the rising unease about the en-

forcement procedures that have taken place under the Martin

Act. In Part II, I shall canvass anew the early history of the

Martin Act and its relationship to the common law of fraud

and both the federal securities laws and state “blue-sky” laws,

with which it must be compared. In Part III, I shall examine

the early historical application of the Martin Act in order to set

the stage for a closer examination of two controversial con-

texts: the prosecution of alleged commercial frauds in the

cases brought against Maurice “Hank” Greenberg of AIG, and

global warming in connection with the now-halted investiga-

tion of ExxonMobil and the recent public nuisance cases

against five oil companies in both California and New York. A

1. N.Y.

. B

. L

§§ 352-c, 353 (McKinney 1996).

2. Securities Act of 1933, 15 U.S.C. § 77a (2012); Securities Exchange

Act of 1934, 15 U.S.C. § 78a (2012).

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 807

brief conclusion argues that the procedural changes that limit

the discretion of the New York Attorney General are more im-

portant, especially in high profile cases, than any substantive

reform of the Martin Act, which on most critical points tracks

the federal securities laws.

NLEASHING OF THE

ARTIN

The Martin Act was passed in 1921, and has been continu-

ously enforced for close to 100 years. But in the last 25 years, it

has been transformed from a relatively modest operation into

an 800-pound gorilla. The transformation started under now-

disgraced former New York Attorney General (NYAG) Eliot

Spitzer, who served as Attorney General between 1999 and

2006. Spitzer ratcheted up his powers under the Martin Act to

cudgel favorable settlements from large firms doing business

in New York. Spitzer’s efforts were often directed against Wall

Street firms that were subject to direct regulation by the Secur-

ities and Exchange Commission (SEC). But using his formida-

ble investigative powers under the Martin Act, Spitzer ex-

tracted large settlements against brokerage houses for finan-

cial fraud. In many cases, the substantive results appear

unexceptionable; he extracted, for example, a huge settlement

from Merrill Lynch because it touted shares to the public that

it had disparaged privately.

But the means he deployed under

the Martin Act were troubling. Spitzer’s operations depended

on his ability to unilaterally conduct major investigations

backed by his power of indictment. Once filed, an indictment

could prove a death knell to financial firms. An indictment

could cost these firms both their licenses and reputations. In

an effort to stave off that threat, firms yielded to various de-

mands, including oversight, that were far more onerous than

the relatively modest fines that could be imposed after a con-

viction was obtained in a legal action, where the defendant

had all the procedural protections of the criminal law. It was

for circumventing of structural and procedural safeguards that

Spitzer was strongly castigated by business groups such as the

Chamber of Commerce, whose president, Thomas Donahue,

denounced Spitzer for appointing himself “investigator, prose-

3. See, e.g., Rebecca Leung, The Sheriff of Wall Street,

CBS N

EWS

(May 23,

2003), http://www.cbsnews.com/news/the-sheriff-of-wall-street/.

808 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

cutor, judge, jury and executioner” over all U.S. businesses.

“Spitzer’s approach,” he claimed, “is to walk in and say, ‘we’re

going to make a deal, and you’re going to pay $600 million to

the state, and you’re going to get rid of this person and that

person, and if you don’t do it by tonight, we’re going to indict

the company.’”

Former NYAG Eric Schneiderman has brought the Martin

Act back into the spotlight. His most notable endeavor was the

climate-change investigation of Exxon Mobil, which involved

activities going back decades on the energy and environment

frontier.

The gist of the charges was that internal politics

turned a company that had been a leader in the fight against

global warming into one that consciously sought to obfuscate

the work of the United Nation’s Intergovernmental Panel on

Climate Change (IPCC).

When these charges failed to pan

out, Schneiderman shifted his investigation from Exxon Mo-

bil’s past conduct to its current efforts to influence the policy

debate over the use of fossil fuels in the light of the risks of

global warming, which in turn led the company to overvalue

its reserves in ways that he claimed presented a false picture of

the company’s financial health.

Schneiderman sought exten-

sive discovery, and changed his theory of the case multiple

times, all within the broad confines of the Martin Act.

His

campaign mirrored Spitzer’s strong-arm tactics but extends

4. U.S Group is Taking on Spitzer,

N.Y. T

IMES

(

Jan. 6, 2005), http://

www.nytimes.com/2005/01/06/business/worldbusiness/us-group-is-taking-

on-spitzer.html.

5. Id.

6. See John Schwartz, Exxon Mobil Climate Change Inquiry in New York

Gains Allies,

N.Y. T

IMES

, Mar. 29, 2016, at B3 (listing the Martin Act as among

Attorney General Schneiderman’s New York-specific tools to investigate Ex-

xon).

7. For a collection of posts attacking the overall pattern of Exxon’s

changing position, see Neela Banerjee, Lisa Song & David Hasemyer, Exxon:

The Road Not Taken,

NSIDE

LIMATE

EWS

(Sept. 16, 2016) [hereinafter The

Road Not Taken], https://insideclimatenews.org/content/Exxon-The-Road-

Not-Taken.

8. John Schwartz, Exxon Mobil Fraud Inquiry Said to Focus More on Future

Than Past,

N.Y. T

IMES

, Aug. 19, 2016, at B1 (in which Schneiderman claims

that, since burning even a fraction of declared oil reserves would heat the

world to a dangerous level, Exxon is overvaluing its oil reserves because the

company will have to leave much of the oil in the ground).

9. See generally People v. PricewaterhouseCoopers, 2016 NYLJ LEXIS

3688, at *2–7 (N.Y. Sup. Ct. 2016).

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 809

them beyond the traditional context of financial fraud. It

makes some sense for a prosecutor to look at sales and promo-

tions to identify fraud against customers to whom the sales

pitches were directed, but it makes little sense to look to a

company’s position on global warming, where a host of other

factors are at play.

At the outset, therefore, it is important to stress one simi-

larity that links these two campaigns—namely, the inordinate

power that it vested in one person, the NYAG, to launch major

attacks against established businesses. In these cases, the process

issues are as important as the substantive ones. It is possible to

engage in prosecutorial abuses against firms that are guilty of

some unrelated offense, since the public may too hastily con-

clude that these shortcuts were justified because they allowed

virtuous prosecutors to attack bad commercial actors. But the

hasty conclusion that the ends justify the means is wrong in

this context, as in so many others, for three reasons. First, the

same dubious means can be used against the innocent as well

as the guilty, bludgeoning them into costly and unjust settle-

ments. Second, bullying tactics undermine the public legiti-

macy of prosecutions against the guilty. Demoralization about

the law-enforcement process leads to the loss of confidence in

the overall system, leading to widespread dissatisfaction both

within the business community and the public at large. Finally,

aggressive prosecution leads firms that operate on the margins

of the law to take excessive precautions, abstaining from sensi-

ble and desirable projects for fear of an unfortunate exposure

to undeserved legal sanctions.

In dealing with both civil and criminal sanctions, there

should be no free pass to the aggressive prosecutor. It is not

just under-deterrence that causes serious social dislocations.

Over-deterrence can, and does, bring about the same result. In

this context it is good to recall the words of Justice Felix Frank-

furter in McNabb v. United States

A democratic society, in which respect for the dignity

of all men is central, naturally guards against the mis-

use of the law enforcement process. Zeal in tracking

down crime is not in itself an assurance of soberness

of judgment. Disinterestedness in law enforcement

does not alone prevent disregard of cherished liber-

10. McNabb v. United States, 318 U.S. 332 (1943).

810 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

ties. Experience has therefore counseled that safe-

guards must be provided against the dangers of the

overzealous, as well as the despotic. The awful instru-

ments of the criminal law cannot be entrusted to a

single functionary. The complicated process of crimi-

nal justice is therefore divided into different parts, re-

sponsibility for which is separately vested in the vari-

ous participants upon whom the criminal law relies

for its vindication.

Frankfurter’s anxiety about excessive reliance on a single

functionary applies with even greater force to many modern

regulatory schemes, of which the most notable is the Con-

sumer Financial Protection Bureau (CFPB), which was created

as part of the 2010 Dodd–Frank Act.

The distinctive features

of CFPB are numerous. The CFPB has enormous powers that

consolidate the enforcement of some 19 different consumer

protection statutes that run the gamut: student loans, credit

cards, home mortgages, and much more. Inside the CFPB, the

complete oversight for its far-reaching operations is lodged in

a single director whose decisions are subject at most to a

highly limited review. There is no budgetary constraint by

Congress, as the agency has by design a guaranteed source of

funding through the Federal Reserve Bank, which cannot be

checked by Congress or, indeed, even the Federal Reserve it-

self. Finally, Dodd–Frank provides that the CFPB Director

serves for a five-year term—which necessarily allows one presi-

dent to bind his successor, given that the Director is subject to

presidential removal only limited grounds—namely for “ineffi-

ciency, neglect of duty, or malfeasance in office.”

The question of whether this unique statutory arrange-

ment offends the basic constitutional structure dealing with

separation of powers on the one side and checks and balances

on the other came to a head in PHH Corporation v. Consumer

Financial Protection Bureau,

under facts that could be scarcely

more illustrative of the dangers of this top heavy organization.

11. Id. at 343 (emphasis added).

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

L. No. 111-203, 124 Stat. 1376 (2010).

13. 12 U.S.C. § 5491(c)(3) (2018).

14. PHH Corp. v. Consumer Fin. Prot. Bureau, 839 F.3d 1 (D.C. Cir.

2016), vacated and reh’g en banc granted, No. 15-1177 (D.C. Cir. 2017).

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 811

PHH specializes in home mortgage lending, and it also partici-

pates in a mortgage insurance market, through its “captive”

subsidiary Atrium Insurance Corporation that supplied rein-

surance to outside insurers that wrote coverage for mortgages

issued by PHH. There was a complex question of whether

PHH was, in accordance with historical practice, exempt from

certain kickback or self-dealing provisions which had pro-

tected it under the Real Estate Settlement Procedures Act,

passed in 1974, so long as the fees paid the mortgage insurers

was reasonable. But once the enforcement functions were

transferred to the CFPB, it ruled, without any change in the

substantive law, that this long-standing interpretation no

longer held. On appeal to the Director, the initial fine (for

this dubious act of statutory interpretation) was increased

from $6 million to $109 million

—a punitive fine after a uni-

lateral reinterpretation of an established interpretation of a

statute on which there was extensive industry reliance.

The imperial nature of this CFPB decision lay it open to

the charge that this concentrated power violates the general

requirements of separation of powers. In one sense the issue

arises in a second-best world because it assumes that indepen-

dent agencies, whose officers are immune from removal at the

will of the President,

are consistent with the basic constitu-

tional system, even though the Constitution itself contains no

explicit authorization for their use. Thus the process of impli-

cation is necessarily invoked to both define the scope and lim-

its of that exception.

In the initial round, Judge Brett Kavanaugh held for him-

self and Senior Judge Raymond Randolph that independent

agencies, or those whose heads can only be removed for

cause,

have too much power to be led by a single official.

Given that all administrative agencies “pose a significant threat

to individual liberty and to the constitutional system of separa-

tion of powers and checks and balances,” they “have histori-

cally been headed by multiple commissioners, directors, or

15. For a discussion of substantive issue, see Richard A. Epstein, Tyranny

Within the Administrative State,

OOVER

NST

.: D

EFINING

DEAS

(May 31, 2016),

https://www.hoover.org/research/tyranny-within-administrative-state.

16. Humphrey’s Ex’r v. United States, 295 U.S. 602, 603 (1935).

17. Id.

18. PHH Corp., 839 F.3d at 8–9.

812 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

board members who act as checks on one another.”

Given

the CFPB head’s level of unilateral authority, Kavanaugh held

that the constitution requires that the President can remove

such an official at will.

That initial decision was reversed en banc by a divided

court that showed the deep progressive/classical liberal split

on this issue.

The majority decision written by Judge Nina

Pillard took a very different view of the overall matter. In its

view, heading an agency with a sole director and not a multi-

ple panel was its virtue because the streamlined structure was

designed “to imbue the agency with the requisite initiative and

decisiveness to do the job of monitoring and restraining abu-

sive or excessively risky practices in the fast-changing world of

consumer finance.”

In response to the Kavanaugh view that

the issue of removal could not be examined apart from the

other structural provisions, Pillard concluded: “The relevance

of ‘internal checks’ as a substitute for at-will removal by the

President is no part of the removal-power doctrine, which fo-

cuses on executive control and accountability to the public,

not the competing virtues of various internal agency design

choices.”

So long as removal could be properly limited for

cause, in her view the other design features did not matter.

As a matter of recent historical decision, the Pillard posi-

tion draws some support from the long history of deference to

Congress in the design of administrative agencies. But as a

matter of first principle, her entire approach is inconsistent

with the basic conception of separation of powers that drove

the initial constitutional design. These design decisions recog-

nize the inherent trade-offs between going too fast and too

slow. Of course, it is important to be zealous in the pursuit of

wrongdoing. But that tendency that creates the risk of over-

zealous behavior, which was on ample display in PHH, makes

the Kavanaugh approach preferable as a matter of principle.

Indeed, it is just that principle of aggressive concentration of

powers in the hands of a single executive, here the NYAG,

that, even more than its substantive provisions, makes the cur-

19. Id. at 5.

20. Id. at 8.

21. PHH Corp. v. Consumer Fin. Prot. Bureau, 881 F.3d 75 (D.C. Cir.

2018).

22. Id. at 81.

23. Id. at 79.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 813

rent operation of the Martin Act so dangerous. In the hands of

cautious NYAGs, the risk of excess is subject to internal politi-

cal constraints. But under NYAGs like Spitzer and Schneider-

man, surely the opposite is true.

Speaking generally, a sound legal system has to contend

with two kinds of error: both over and underenforcement. It is

always hard to know whether it is socially preferable to trade

away the upside of effective enforcement in order to minimize

the downside of wrongful or excessive punishments. Any effort

to beef up procedural protections will unfortunately trammel

good public officials but simultaneously limit the harm from

zealous overenforcement. In a perfect world, I would prefer

that good leaders be unconstrained by procedural hurdles.

But in our imperfect world, procedural protections are neces-

sary, indeed urgent. The ultimate prudential calculation re-

quires assessing which form of error is greater, and, as Justice

Frankfurter’s admonition reminds us, it is never wise to disre-

gard the downside. But it is just that error that the Martin Act

has committed—especially as it has been interpreted and ap-

plied by recent NYAGs.

Textually, there is nothing about the Martin Act that com-

pelled Schneiderman’s behavior any more than there is in the

Federal Trade Commission Act of 1914, Section 5, which

broadly prohibits “using unfair methods of competition in or

affecting commerce, and unfair or deceptive acts or practices

in or affecting commerce . . . .”

But it takes awareness of the

dangers to control the excesses. Clearly there is wiggle room in

each of these two key statutes, which has led to some highly

controversial cases. Yet ironically, it took a one-page statement

from the FTC—a multimember body—to announce that it

would confine the application of this elastic provision to oper-

ate as an adjunct to the antitrust laws dealing with anticompe-

titive injury instead of a freewheeling statute that could be di-

rected at any contract that displeased the FTC.

Given the

24. 15 U.S.C. § 45(a)(1) (2012).

25.

. T

RADE

OMM

’

, S

TATEMENT OF

NFORCEMENT

RINCIPLES

E-

GARDING

“U

NFAIR

ETHODS OF

OMPETITION

”

UNDER

ECTION

OF THE

FTC

(Aug. 13, 2015). For my account of the evolution, which was largely

attributable to the work of Commissioner Joshua Wright, who has since re-

turned to the George Mason Law School Faculty, see Richard A. Epstein,

When Bureaucrats Do Good,

OOVER

NST

.: D

EFINING

DEAS

(Aug. 17, 2015),

https://www.hoover.org/research/when-bureaucrats-do-good.

814 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

powers vested in the NYAG, some check on his unilateral pow-

ers could go a long way toward curbing prosecutorial excesses

under the Martin Act. Such a procedural change could then

be buttressed substantively by limiting the Martin Act to those

cases that are akin to the common-law tort of deceit.

Unfortunately, all of the implicit safeguards built into the

earlier common-law-inspired use of the Act have been eroded

by unsupported extensions of its substantive reach, which in

turn have increased the danger of invoking the extraordinary

remedial provisions built into the Act in the first place. These

provisions give huge investigative powers to the NYAG, acting

solely on his own authority, long before any trial. For example,

Section 352 of the Act enables the NYAG to undertake confi-

dential investigations in which he can subpoena witnesses,

compel attendance, and examine witnesses under oath, either

in private or before a magistrate court of record.

Section 354

gives the NYAG the unilateral authority to order a public inves-

tigation by filing an application before a trial-court judge, who

has no choice but to grant the application.

To paraphrase

Chief Justice John Marshall, the power to investigate is the

power to destroy,

especially with respect to those businesses

whose reputation is critical to their ability to remain open.

Reputation, of course, is, and should be, of great importance

in the securities business. The decisions of individual and insti-

tutional investors to flee a firm that has committed a serious

violation of industry norms takes its effect immediately and

does so without the need for costly and prolonged litigation. It

is for that reason that losses in share values routinely exceed

any potential liability or fines for illegal behavior.

This bond-

ing mechanism makes good sense when the bad news is accu-

rate about a firm. But it should be a source of unease when the

loss in stock value from a triggering event does not stem from

any actual bad conduct by the firm in question. The risk of

social harm from over-deterrence is exceedingly high when-

ever the simple announcement of an investigation can be

26. N.Y.

. B

§ 352(2) (2016).

27. Id. at § 354.

28. See McCulloch v. Maryland, 17 U.S. 316, 431 (1819) (“[T]he power to

tax involves the power to destroy . . . .”).

29. Jonathan M. Karpoff, D. Scott Lee & Gerald S. Martin, The Cost to

Firms of Cooking the Books, 43

J. F

. & Q

UANTITATIVE

NALYSIS

547

, 582

(2008).

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 815

enough to trigger a suspension of the licenses needed to do

business—not to mention follow-on consequences in stock

price. In those settings, there is intolerable pressure to settle a

case short of the initiation of suit.

The federal government enjoys similar forms of leverage

under the federal securities laws, which, to some extent, ex-

hibit the same flaws as the Martin Act.

This excessive concen-

tration of power is exemplified by the practice of the SEC to

try its own cases before administrative law judges (ALJs) that

the SEC appoints itself. Concentrating that power in the hands

of a political commission whose membership is determined

along party lines is asking for trouble. One recent decision on

the topic, Bandimere v. SEC,

held that the ALJs appointed by

the SEC wield sufficient power in individual cases to count as

inferior officers under the Appointments Clause. Therefore,

the Bandimere court held that the ALJs would need to be ap-

pointed, without Senate confirmation, either by the President,

the Courts of Law, or the Head of Department. As the SEC was

none of the above, its commissioners could not constitution-

ally appoint ALJs.

The District of Columbia Circuit in Ray-

mond J. Lucia Cos. v. SEC

disagreed with this reading, and

held that the ALJs should be treated only as government em-

ployees because the decision of the ALJ was not final, but ulti-

mately depended for its validity on an enforcement decision

by the Commission as a whole, which has the power to review

every case. That decision let the SEC resort to deferred prose-

cution agreements, which allowed parties to retain their li-

censes only if they acceded to onerous behavioral and struc-

tural changes in the conduct of their business.

30. For a discussion of the perverse incentives that arise when the costs of

being indicted exceed the fines that could be imposed on conviction, see

Richard A. Epstein, Deferred Prosecution Agreements on Trial: Lessons from the

Law of Unconstitutional Conditions, in

ROSECUTORS IN THE

OARDROOM

: U

SING

RIMINAL

AW TO

EGULATE

ORPORATE

ONDUCT

38 (Anthony S. Barkow &

Rachel E. Barkow eds., 2011).

31. Bandimere v. SEC, 844 F.3d 1168, 1179 (10th Cir. 2016), reh’g en banc

denied, 855 F.3d 1128 (10th Cir. 2017).

32. Id.

33. Raymond J. Lucia Cos. v. SEC, 832 F.3d 277 (D.C. Cir. 2016).

34. See, e.g., James R. Copland & Rafael Mangual, The Shadow Regulatory

State at the Crossroads: Federal Deferred Prosecution Agreements Face an Uncertain

Future,

ANHATTAN

NST

(June 27, 2017), http://www.manhattan-institute.

org/html/shadow-regulatory-state-crossroads-federal-deferred-prosecution-a

816 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

For the moment at least, these abusive practices have

come to a halt as Justice Elena Kagan upended the District

Court decision by holding in Lucia v. SEC that the ALJs that

were appointed by the Securities and Exchange Commission

counted as inferior officers under the Appointments Clause.

Thus, they could not be appointed by the Commission’s staff,

but had to be appointed by its head alone.

The court relied

heavily on its earlier decision in Freytag v. Commissioner, which

held that the special trial judges in the United States Tax

Court were inferior offices because, even though they were not

authorized under statute to issue final decisions, the extensive

scope of their authority meant that they had sufficient power

to shape the overall scope of the decision.

The Appointments Clause issue is, however, only the tip

of a larger due process iceberg.

The decision in Lucia spent

no time at all discussing the particulars of the case, which were

deeply troubling from a due process point of view. Raymond

Lucia was permanently barred from practice as an investment

advisor because the administrative law judge in the case, Cam-

eron Elliot, found him guilty of serious violations of the securi-

ties laws for giving his “Buckets of Money” presentation to a

general audience. Under the strategy presented by Lucia, an

investment portfolio would contain both stocks and bonds.

The low-risk investments would be sold first, while the high-

risk investments would be allowed to appreciate. Lucia’s large

slide deck contained hypothetical examinations of how his

method works, and these were deemed to be materially mis-

leading and thus in violation of the anti-fraud provisions of the

Investment Advisors Act of 1940

and applicable regula-

tions.

This was despite the fact that the slides had passed

greements-face-uncertain. For my longstanding opposition to these agree-

ments, see Richard A. Epstein, The Dangerous Incentive Structures of Nonprosecu-

tion and Deferred Prosecution Agreements,

ERITAGE

OUND

(June 26, 2014),

http://www.heritage.org/research/reports/2014/06/the-dangerous-incen

tive-structures-of-nonprosecution-and-deferred-prosecution-agreements.

35. Lucia v. SEC, No. 17-130, slip op. at 5 (U.S. June 21, 2018).

36. Freytag v. Comm’r, 501 U.S. 868, 891 (1991).

37. For my longer discussion of this issue, see Richard A. Epstein, Miscar-

riage of Justice at the SEC,

OOVER

NST

.: D

EFINING

DEAS

(Apr. 30, 2018),

https://www.hoover.org/research/miscarriage-justice-sec.

38. 15 U.S.C. §§ 80b-1 to 80b-21 (1940).

39. See 17

C.F.R.

§ 275.206(4)-1(a)(5) (2012) (prohibiting misleading ad-

vertising).

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 817

muster before a private industry self-regulatory organization

known as FINRA (The Financial Industry Regulatory Author-

ity), on which the U.S. securities regulation regime relies heav-

ily for enforcement.

In addition, the SEC itself raised no ob-

jections to the slides when it previewed the presentation.

None of the 50,000 people who attended these presentations

over the years filed any complaint against Lucia, let alone lost

any money as a result. At these presentations, Lucia asked at-

tendees to contact him directly if they wanted additional per-

sonal advice.

The case was then referred to ALJ Cameron, who had

never previously ruled against the SEC’s enforcement division

in the approximately 50 or so cases he oversaw. According to

Professor Ronald Mann, Elliot “adopted a bright-line rule of

issuing lifetime bans on employment in the investment indus-

try against any defendants who had the nerve to contest the

proceedings against” the SEC.

This case followed true to

form. Elliot presided over a full-scale trial, after which he con-

cluded that Lucia should be suspended for life from the secur-

ities profession. This initial decision was not final; it had to be

reviewed by the agency’s five commissioners, who in 2015 fol-

lowed Elliot’s recommendation in a 3–2 decision.

In the

Court’s opinion, Justice Kagan simply prefaced her remarks by

noting that the Dodd–Frank Act allows the Commission to pre-

side over these cases, but also allows it to delegate that activity

to an ALJ.

Her opinion never raised these due process issues

despite the fact that the Commission’s power to select individ-

ual judges on an ad hoc basis gives rise to the appearance of

bias. Such risk is hardly cured even if the SEC mends its ways

and now requires the head of the SEC to make those appoint-

ments. It should seem clear that the risk of bias is not con-

40. See generally FINRA, https://www.finra.org (last visited Aug. 11, 2018).

41. Ronald Mann, Argument Preview: Justices Turn to Constitutional Limits on

Appointments of Administrative Law Judges,

SCOTUS

BLOG

(Apr. 16, 2018), http:

//www.scotusblog.com/2018/04/argument-preview-justices-turn-to-constitu

tional-limits-on-appointment-of-administrative-law-judges/.

42. For the opinion and order of the Commission, see Lucia, Exchange

Act Release No. 34-75662, 2015 WL 5172953 (Aug. 11, 2015). An interim

remand order is unreported. For the relevant initial decision of the adminis-

trative law judge, see Initial Decision Release No. 495, 2013 WL 3379719

(July 8, 2013) (ALJ).

43. Lucia v. SEC, No. 17-130, slip op. at 1 (U.S. June 21, 2018).

818 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

fined solely to the Martin Act, but also arises under the SEC

generally, where delegation can result in the same kind of ag-

gregation of power in the hands of a single public official.

These issues will surely have to be vetted in future cases.

In addition to trusting plausibly unconstitutionally ap-

pointed ALJs with immense power, the SEC resorts to deferred

prosecution agreements, whereby the defendant accedes to

prosecutors’ demands for structural and behavioral changes in

exchange for deferring a prosecution that could, among other

things, result in a loss of licenses to do business. These agree-

ments can prove to be especially onerous on defendants, given

that they can be imposed without offering defendants the ben-

efit of any form of judicial oversight.

These two features of

the SEC’s enforcement regime show how easy it is for adminis-

trative agencies to centralize power and then abuse it to the

detriment of private parties.

The Martin Act’s scheme is still more perverse, inasmuch

as it centralizes power in the hands of a single individual. The

need for some judicial or institutional check against self-ag-

grandizement is as important with the SEC as in the Martin act

context, and remains critical even if the substantive law re-

mains relatively unchanged. The level of prosecutorial discre-

tion in all such cases is sufficiently broad that the identity of

the prosecutor really matters. Put in its simplest form, it is bet-

ter to have an overbroad statute with a prudent prosecutor

than a finely tuned statute with an overzealous prosecutor. Af-

ter the fact remedies are notoriously ineffective in controlling

abuses of prosecutorial discretion.

Ex ante schemes of con-

trol are more important. This point becomes even clearer after

reviewing the substantive evolution under the Martin Act, in its

relationship to both the common law and the SEC.

44. See, e.g., Copland & Mangual, supra note 34. For Epstein’s opposition

to these agreements, see Epstein, supra note 34.

45. Peter A. Joy, The Relationship Between Prosecutorial Misconduct and

Wrongful Convictions: Shaping Remedies for a Broken System,

2006 W

. L. R

399

, 427 n.139 (surveying literature to this effect).

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 819

II.

ARTIN

THE

OMMON

-L

CTION FOR

ECEIT

AND THE

EDERAL

ECURITY AND

XCHANGE

CTS

A. The Origins of the Martin Act

To better understand the recent growth of the Martin Act,

it is useful to look back to its origins and basic structure. When

enacted in 1921, the Martin Act was among the last of the

many “blue-sky” laws enacted in the early part of the twentieth

century to deal with a spate of flagrantly dishonest marketing

practices targeting “an enormous population with an unusu-

ally high per capita wealth suddenly made security con-

scious.”

These blue-sky laws were intended to prevent un-

scrupulous promoters from bilking innocent investors out of

their savings by offering them securities whose market value

was no greater than the blue sky above.

Early Supreme Court decisions showed no hesitation in

sustaining the constitutionality of the blue-sky laws, chiefly on

the ground that they fell squarely within the scope of the po-

lice power. Passed at the height of the so-called Lochner era,

when the protection of economic liberties from government

regulation was at its zenith,

these laws produced little consti-

tutional doubt.

The worry that these laws might over-deter

46. Ambrose V. McCall, Comments on the Martin Act, 3

ROOK

. L. R

190,

193 (1933).

47. See Hall v. Geiger-Jones Co., 242 U.S. 539, 550 (1917) (“The name

that is given to the law indicates the evil at which it is aimed; that is, to use

the language of a cited case, ‘speculative schemes which have no more basis

than so many feet of ‘blue sky’; or, as stated by counsel in another case, ‘to

stop the sale of stock in fly-by-night concerns, visionary oil wells, distant gold

mines, and other like fraudulent exploitations.’”). For discussion, see Elisa-

beth Keller & Gregory A. Gehlmann, Introductory Comment: A Historical Intro-

duction to the Securities Act of 1933 and the Securities Exchange Act of 1934, 49

HIO

. L.J.

329, 331–34 (1988).

48. The era took its name, of course, from Lochner v. New York, 198 U.S.

45 (1905), which struck down a New York law that limited workers in certain

classes of bakeries to 10 hours per day and 60 hours per week, as an uncon-

stitutional “labor” statute that was directed more toward suppressing compe-

tition than protecting health and safety—which indeed was the case. See, e.g.,

AVID

E. B

ERNSTEIN

, R

EHABILITATING

OCHNER

(2011).

49. See Hall, 242 U.S. at 552 (upholding Ohio’s blue-sky law); Caldwell v.

Sioux Falls Stock Yards Co., 242 U.S. 559 (1917) (upholding South Dakota’s

blue-sky law); Merrick v. N.W. Halsey & Co., 242 U.S. 568 (1917) (upholding

Michigan’s blue-sky law).

820 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

legitimate business behavior was a distant worry in the hectic

years of the First World War, when crooked behavior was ram-

pant.

Since its original passage, the scope of the Martin Act has

been expanded by legislative action. In 1955, Section 352-c au-

thorized the Attorney General to institute criminal proceed-

ings against parties engaged in fraudulent practices, even ab-

sent proof of scienter or intent.

In 1976, Section 353(3) au-

thorized the NYAG to seek monetary restitution for persons

defrauded by the defendant’s activities.

The major doctrinal innovation of the Martin Act was to

soft-pedal the role of scienter, reliance, and damages—all es-

sential elements of a common-law fraud claim. However,

within the context that sparked passage of the Martin Act,

these shortcuts were perfectly appropriate responses for deal-

ing with blatant fraudsters, and the Martin Act preserved other

common-law rules that kept liability from running rampant.

For instance, its application extended to misrepresentation by

omissions as well as by affirmative utterances, but was limited

to material misrepresentations, ruling out mere statements of

opinions and predictions. The Martin Act also respected the

common-law privity limitations that grew up in connection

with the tort of deceit, which blocked litigation for persons

who suffered only from the indirect consequences of fraud.

There is much that is distinctive about the Martin Act, but

there is also much that it shares with both federal securities

law and the older common law. As the New York Court of Ap-

peals has recognized, “[a]lthough the Martin Act was enacted

in 1921, its present form generally tracks the Federal securities

acts of 1933 and 1934. Accordingly, we have looked to Federal

50. In those years, “[s]windlers stalked Gotham’s streets, fleecing the

people who were investing with solo speculators and putting money into the

stock market for the first time in a burst of postwar patriotic fervor.”

Nicholas Thompson, The Sword of Spitzer,

EGAL

(May/June 2004),

https://www.legalaffairs.org/issues/May-June-2004/feature_thompson_may

jun04.msp.

51. See People v. Landes, 645 N.E.2d 716, 717–18 (N.Y. 1994) (outlining

this history).

52. See Assured Guar. Ltd. v. J.P. Morgan Inv. Mgmt., Inc., 962 N.E.2d

765, 768 (N.Y. 2011) (describing this history); N.Y.

. B

§ 353(3)

(McKinney 2014).

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 821

court decisions construing those statutes when interpreting

our own.”

The original Martin Act gave a broad definition to the

types of activities that fell within its scope. Section 352-c con-

tains a list of activities prohibited by the statute:

It shall be illegal and prohibited for any person, part-

nership, corporation, company, trust or association,

or any agent or employee thereof, to use or employ

any of the following acts or practices:

(a) Any fraud, deception, concealment, suppres-

sion, false pretense or fictitious or pretended

purchase or sale;

(b) Any promise or representation as to the fu-

ture which is beyond reasonable expectation or

unwarranted by existing circumstances;

false, where the person who made such repre-

sentation or statement: (i) knew the truth; or (ii)

with reasonable effort could have known the

truth; or (iii) made no reasonable effort to ascer-

tain the truth; or (iv) did not have knowledge

concerning the representation or statement

made;

where engaged in to induce or promote the issuance,

distribution, exchange, sale, negotiation or purchase

within or from this state of any securities or commod-

ities, as defined in section three hundred fifty-two

[giving extensive investigative powers to the Attorney

General] of this article, regardless of whether issu-

ance, distribution, exchange, sale, negotiation or

purchase resulted.

The initial version of the law had a schizophrenic quality,

being broad in scope but feeble in its application. Indeed, the

statute has been described as “deliberately enfeebled” at

birth.

As Nicholas Thompson explained in criticizing

Spitzer’s ramped-up enforcement efforts, the original Martin

Act embodied a political compromise whereby the broad

reaches of the law were directed to marginal players, both in-

53. See Landes, 645 N.E.2d. at 718.

54.

N.Y. G

. B

. L

§ 352-c(

(McKinney 1982).

55. See Thompson, supra note 50.

822 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

side and outside the financial industry, while the established

players received something of a free pass.

For instance,

Thompson noted that the New York law did not have a regis-

tration requirement of the sort found in many other blue-sky

statutes, because for large institutions “a registration law was a

bureaucratic burden to be avoided. A simple fraud statute

seemed like a good way to swat down small-time sharks and

keep the field open for themselves.”

Enforcement in this

early era was predictably light.

But the shifting tides of en-

forcement notwithstanding, a comparison with the common-

law tort of fraud or deceit shows how wide a liability net the

Martin Act has always cast.

B. The Common Law of Deceit

To put the Martin Act in context, it is useful to compare

its provisions to common-law tort of deceit that awarded pri-

vate parties money damages for the loss attributable to fraudu-

lent conduct.

The elements of the common-law tort are em-

bodied in Section 525 of the Restatement (Second) of Torts.

One who fraudulently makes a misrepresentation of

fact, opinion, intention or law for the purpose of in-

ducing another to act or to refrain from action in re-

liance upon it, is subject to liability to the other in

deceit for pecuniary loss caused to him by his justifia-

ble reliance upon the misrepresentation.

It is also instructive to note that the standard federal defi-

nition of securities fraud takes exactly the same tack. Under

federal law, plaintiffs must prove “(1) a material misrepresen-

tation or omission by the defendant; (2) scienter; (3) a con-

nection between the misrepresentation or omission and the

purchase or sale of a security; (4) reliance upon the misrepre-

56. Id.

57. Id.

58. Lydie N.C. Pierre-Louis, Hedge Fund Fraud and the Public Good, 15

ORDHAM

J. C

ORP

. & F

. L.

21, 34 (2009) (“The original Martin Act was a

weak law and no enforcement actions were commenced for years after its

adoption.”).

59. See Pasley v. Freeman (1789) 100 Eng. Rep. 450 (KB).

60.

ESTATEMENT

ECOND

)

ORTS

§ 525 (A

. L

. I

NST

. 1965).

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 823

sentation or omission; (5) economic loss; and (6) loss causa-

tion.”

The tort and its constitutive elements necessarily give rise

to all sorts of knotty doctrinal questions. What is the basis of

liability? Does the tort cover fraud only, or does it extend fur-

ther? How is the element of causation cashed out when the

loss is financial, as opposed to the more familiar tort context

of physical injury? Is there a materiality limitation? Does liabil-

ity for false statements cover omissions, mere opinions, predic-

tions, and statements of intention? The express text of the

Martin Act answers some of these questions, but it remains si-

lent on other key issues. It is therefore common for cases

under the Martin Act to draw explicit comparisons to the fed-

eral security acts, which in turn rely on explicit comparisons to

the various common-law rules. All of these interactions must

be examined.

C. Basis of Liability: Fraud, Recklessness, Negligence, and Beyond

At common law, the basic instinct was to impose liability

for deceit but not for innocent or even negligent misrepresen-

tations.

There are good reasons for this distinction, even

though the harm to the aggrieved party may be the same. Im-

perfect information—from whatever source—distorts any deci-

sion-making process, because it leads people to attach false

weights to the goods and services they purchase or sell. Attach

too high a value to some object, and you will pay too much for

it. Attach too low a value, and you will part with it for too low a

price.

Unfortunately, incomplete information is the normal

state of human affairs, so any legal system must be picky about

the shortfalls in information that it will target for attack. The

basic presumption that all individuals should ordinarily bear

the costs of their own mistakes provides a sensible point of de-

parture for two reasons. First, it offers a strong incentive for

individuals to ferret out good information to counteract the

61. Stoneridge Inv. Partners v. Scientific-Atlanta, 552 U.S. 148, 157

(2008). See also Pits, Ltd. v. Am. Express Bank Int’l, 911 F. Supp. 710, 719

(S.D.N.Y. 1996) (noting the exact parallel between the common-law action

for deceit in New York and an action for federal securities fraud under Sec-

tion 10(b) of the Exchange Act and SEC Rule 10b-5).

62. See, e.g., Derry v. Peek [1889] 14 App. Cas. 337 (HL).

824 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

bad, so long as the incremental value of superior information

outweighs the cost in time, effort, and money it would take to

learn it. Second, this rule protects other parties by removing

(in most cases) the obligation to correct the errors of their

trading partners, some of which they may not know at all. The

security of transactions is remarkably improved if everyone un-

derstands that his own mistakes are not a ground on which he

can upset transactions made with others.

This rule holds best where there are unilateral mistakes

for which no one else is to blame, let alone one’s trading part-

ner. Like all legal rules, however, this one is subject to excep-

tion. More specifically, fraud in the inducement (note the

causal element) changes the calculation because now one per-

son has invested in disinformation of another solely for his

own advantage. In general, it is common to treat reckless disre-

gard of the truth as though it were a specimen of fraud, on the

ground that the parties who engage in these steps well know

what they have done wrong.

This is a far cry from both inno-

cent misrepresentation and even negligent misrepresenta-

tion,

neither of which involve conscious awareness of the

likelihood of falsehood.

This carefully crafted definition of fraud makes sense

functionally because the correct legal approach is to always

start with the low-hanging fruit: the area where the minimum

of legal intervention will generate the maximum of social im-

provement. Under this simple test, the primacy of fraud—

which is the primary target of the common law, the federal

securities laws, and New York’s Martin Act—should be appar-

ent. On the one hand, fraud constitutes the greatest peril to

voluntary transactions, because fraudsters will work overtime

to deceive the individuals from whom they wish to extract

money or other forms of private benefit. On the other hand,

fraud is far less common than simple mistake or ordinary neg-

ligence, so both civil and criminal enforcement may properly

focus on the most egregious form of misconduct. The dangers

of fraud are recognized in the common-law rules that govern

its application, which, in virtually every jurisdiction, specifically

63. See id.

64. See id. at 375 (“Fraud is proved when it is shown that a false represen-

tation has been made (1) knowingly, or (2) without belief in its truth, or (3)

recklessly, careless whether it is true or false.”).

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 825

provide that contributory negligence is not a defense to a

fraud action, precisely because we want people to be able to

know that they are protected from deliberate fraud when they

rely on information received from others.

To be sure, there

remains some residual risk that people will make mistakes be-

cause of the innocent misrepresentations of others. But in

most contexts, where there is a parity of access of information

to both parties, the more prudent rule is to eliminate liability

altogether, cautioning both sides to confirm the key assump-

tions on which they base their own decisions.

Negligent misrepresentations present a harder case, given

that it is never clear on whom the burden of discovering the

hidden truth of various assertions should fall. There has long

been a reluctance to develop any generalized remedy of negli-

gence that parallels the scope of fraud. The usual legislative

efforts are to impose these duties of care only in those cases in

which good institutional reasons make it clear that the parties

on one side of a transaction (say, the issuer of a new class of

securities) are likely to have better and cheaper access to infor-

mation than those on the other (such as the multitude of in-

vestors interested in buying the securities). In Britain, one

early law reflecting this compromise was the Director’s Liabil-

ity Act of 1890,

passed in the immediate aftermath of Derry v.

Peek, which made directors and promoters liable to purchasers

of stocks and bonds unless the director or promoter could

show that “he had reasonable ground to believe,” and at all

material times did believe, his statements to be true.

Modern

American law goes one step further and holds firms strictly lia-

ble for false statements made in a prospectus.

Nonetheless, in most contexts other than the highly spe-

cialized one of securities sales, a party can bear the risk of ordi-

65. See Seeger v. Odell, 115 P.2d 977, 980 (Cal. 1941) (“Negligence on

the part of the plaintiff in failing to discover the falsity of a statement is no

defense when the misrepresentation was intentional rather than negli-

gent.”).

66. Director’s Liability Act 1890, 53 & 54 Vict. c. 64 (Eng.).

67. Id. § 3(a).

68. 15 U.S.C. § 77k(a) states:

In case any part of the registration statement, when such part be-

came effective, contained an untrue statement of a material fact or

omitted to state a material fact required to be stated therein or

necessary to make the statements therein not misleading, any per-

son acquiring such security . . . [may] sue.

826 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

nary negligence by his trading partner, in no small part be-

cause the means of self-protection are readily available. One

simple line of defense is to get independent information from

two or more people, and then to push harder when there is a

discrepancy between them. The great methodological mistake

on this topic is to assume that the question of liability for negli-

gence should be resolved as one of abstract theory, when in

fact the more reliable method is to look at the explicit terms of

liability in this context, as in so many others, where the risk of

disproportionate loss is one that leads to routine exemptions

from liability for negligence.

Thus, it is common for brokers

and other suppliers of information to demand (and receive)

explicit protection against liability for losses resulting from

merely negligent counsel. The alternative position, treating

the advice as an implicit guarantee, could lead to the break-

down of the industry, because it would allow the customer to

keep the upside of successful trades while gaining the right of

action against the broker in the event of poor outcomes. No

party would pay for this kind of insurance in a voluntary mar-

ket, which is why it is routinely not required as a matter of law.

In other contexts, liability for a wholly innocent misrepresenta-

tion will sometimes be less important if the defendant who rec-

ommends an investment to the plaintiff has made that same

investment himself. After all, it is very common for people who

seek to recruit others into some risky venture to announce at

the outset that they are putting their own money at risk as well,

which acts as an implicit bond against poor advice. And once

the promoter’s own private investigations reveal the true state

of affairs, we no longer deal with innocent misrepresentations

but with ordinary fraud.

The plain text of the Martin Act does not compel grave

departures from the common law’s sensible distinctions. The

Act imposes liability for any false statement connected with se-

69. Hedley, Byrne & Co. v. Heller & Partners, Ltd. contains a powerful de-

nunciation of the various distinctions between paid and gratuitous transac-

tions before switching gears and honoring the standard exemption clause

put into the contract, by saying: “A man cannot be said voluntarily to be

undertaking a responsibility if at the very moment when he is said to be

accepting it he declares that in fact that he is not.” [1964] AC 465 (HL) 533.

The same rejection of negligence applies in cases where there is an effort to

hold someone responsible in ordinary negligence for some catastrophic loss

that might have been prevented by an adequate inspection.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 827

curities trading when the party making the statement “(i)

knew the truth; or (ii) with reasonable effort could have

known the truth; or (iii) made no reasonable effort to ascer-

tain the truth; or (iv) did not have knowledge concerning the

representation or statement made.”

If the statute were meant

to institute a strict-liability rule, there would be no need for

this careful enumeration of mental states, and the text does

not support a strict-liability reading. The first case described is

a clear case of intent, and the second and third refer to negli-

gence. The fourth mental state is the hardest to construe, and

could admittedly be bent into a strict-liability shape, but the

better reading is that it targets statements made by parties who

know they lack the knowledge they pretend to have.

Judicial interpretation, however, has diverted the Act

from this sensible reading, and it is now commonly said that

the Martin Act requires no proof of scienter in civil actions.

In an early case interpreting the Act, the New York Court of

Appeals urged: “The words ‘fraud’ and ‘fraudulent practice,’

in this connection, should therefore be given a wide meaning,

so as to include all acts, although not originating in any actual

evil design or contrivance to perpetrate fraud or injury upon

others, which do by their tendency to deceive or mislead the

purchasing public come within the purpose of the law.”

More recently, in People v. Sala, a New York appellate court

stated that “the People need only prove that the defendant

committed an intentional act constituting fraud, which under

the Martin Act ‘includes all deceitful practices contrary to the

plain rules of common honesty and all acts tending to deceive

or mislead the public.’”

And in State v. 7040 Colonial Road

70. N.Y.

. B

§ 352-

(1)(

71.

AROLD

K. G

ORDON

ONES

NFORCEMENT

ROCEEDINGS

NDER

ORK

’

ARTIN

26 (2015), https://www.jonesday.com/files/Publi

cation/cc6cfc9e-1517-4707-958d-8ea80d2042c2/Presentation/PublicationAt

tachment/9be275fc-e882-499c-9045-9a429519ab82/FebMar15_%20NYSup

plement_MartinActFeature.pdf; John C. Coffee, Jr., On Thin Ice: Climate

Change, Exxon, the NYAG and the Martin Act,

CLS B

LUE

LOG

(Nov. 23,

2015), http://clsbluesky.law.columbia.edu/2015/11/23/on-thin-ice-climate-

change-exxon-the-nyag-and-the-martin-act/.

72. New York v. Federated Radio Corp., 154 N.E. 655, 657 (N.Y. 1926).

73. People v. Sala, 695 N.Y.S.2d 169, 177 (App. Div. 1999) (citing

AUF-

MAN

, P

RACTICE

OMMENTARIES

, M

INNEY

’

ONS

AWS OF

N.Y., B

OOK

19,

ENERAL

USINESS

AW ART

23-A

, at 32-33); see also People v. Lexington Sixty-

First Assocs., 345 N.E.2d 307, 311 (N.Y. 1976).

828 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

Associates Co., a New York trial court held that the term “fraud”

as used in the Martin Act “includes all deceitful practices con-

trary to the plain rules of common honesty and all acts tending

to deceive or mislead the public, whether or not the product

of scienter or intent to defraud.”

These expansive definitions play fast and loose with the

distinction between “deceive” and “mislead.” In so doing, they

create the risk of imposing ill-considered liability on innocent,

but false, statements. But that perceived risk is often greater in

theory than in practice, as the facts of many of the landmark

Martin Act cases illustrate. The broad statement in Sala was

made in connection with a criminal prosecution for a fraudu-

lent plan to gain tenant approval for a conversion of rental

units to condominiums under New York City’s convoluted

Rent Stabilization Law.

The landlord filled vacant units with

family and friends in order to boost the number of unit hold-

ers who voted in favor of the conversion.

This elaborate

scheme was plainly an intentional—indeed, a conspiratorial—

deception. Similarly, 7040 Colonial Road involved a successful

suit to enjoin the conversion of residential premises into a co-

operative arrangement, when the Attorney General deter-

mined that the defendants had not disclosed certain key facts

relating to the financial condition of the premises, including

its own failure to make good on the maintenance obligations

for the retained units.

This case, too, looked like a culpable

misrepresentation (albeit one by omission, a wrinkle discussed

below). Neither of these real estate transactions involved a

wholly innocent misrepresentation, and both were sensibly

covered by Section 352-e(1) of the Martin Act.

Thus, for the archetypal Martin Act case of yore, the eli-

sion of the scienter issue was more troubling in theory than in

practice. Indeed, the New York courts’ frequent invocation of

“deceit” and the standard of “common honesty” could be char-

74. State v. 7040 Colonial Rd. Assocs. Co., 671 N.Y.S.2d 938, 941 (Sup. Ct.

1998) (citing

AUFMANN

, I

NTRODUCTION AND

OMMENTARY

VERVIEW

IN-

NEY

’

ONS

. L

AWS OF

ORK

, B

OOK

19, G

ENERAL

USINESS

, A

RTICLE

23-A,

at 32–33).

75. Sala, 695 N.Y.S.2d at 177; N.Y.C.

DMIN

. C

ODE

§§ 26-501 to 26-520

(2017).

76. Lexington Sixty-First Assocs., 345 N.E.2d at 311.

77. 7040 Colonial Rd. Assocs. Co., 671 N.Y.S.2d at 941–42.

78. N.Y.

. B

352(1)

(McKinney 2014).

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 829

itably read to parallel the common law’s sensible hesitance to

impose liability for innocent and negligent misstatements, the

cases’ no-less-frequent disclaimers of any scienter requirement

notwithstanding. At the very least there is a stipulated element

of the definition of fraud, given that Section 353(1) of the

Martin Act authorizes actions by the Attorney General against

“practices or transactions heretofore referred to as and de-

clared to be fraudulent practices.”

Moreover, the NYAG’s ef-

fort to obtain only injunctive relief minimized the importance

of the intent issue: it is perfectly proper to enjoin harmful be-

havior, regardless of the defendant’s state of mind. A strict-

liability rule works well for enjoining ordinary nuisance cases,

and those rules can apply just as well in these financial situa-

tions. Indeed, once an injunction is sought, no one could

claim, having been warned, that the activities in question re-

mained unintentional, no matter how innocently they began.

D. Statements of Fact: Materiality, Omissions, Opinions, and

Predictions.

In many ways, the single most important feature of the

common-law action for deceit is that it covers only statement

of facts. The term “facts” can obviously be read very broadly to

cover all assertions that a given state of affairs corresponds to

the description that the defendant has made of it. Indeed, in

most fraud cases, the point passes by without dispute. If the

defendant claims that he owns a well-oiled machine that turns

out not to function at all, it does not take deep philosophical

awareness to conclude that this statement of fact is false, and

therefore imposing civil recovery and criminal punishment is

so clear that no one gives the matter a second thought. In a

hardcore fraud case, lying does not consist only in shaving cor-

ners. It also consists in making up matters out of whole cloth.

Hence the term “blue sky” accurately reflects the proper state

of affairs when the fraud goes to the sale of stock.

Nonetheless, just as the term “statement of fact” impli-

cates some easy cases, it also gives rise to some hard ones. The

use of language is often variable, and there are some state-

ments that are regarded properly as weighty and consequent-

ial and others that are rightly dismissed as light and trivial. The

law of fraud must take this complication into account, and in

79. Id.

830 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

response it develops an initial filter of “materiality” to indicate

which statements are capable of supporting potential liability

and which are too evanescent to matter. A statement’s context

matters as well as its content; sometimes a statement is made in

relative isolation, while in other cases the identical statement

may be just one of many statements, often made by many inde-

pendent individuals, about a given issue. Clearly the statement

is more likely to be material when it stands solo than when it is

part of a symphony of statements concerning broad matters of

public interest where other voices have weighed in.

Subject to that caveat, the overarching inquiry into mate-

riality can conveniently be broken down into three separate

issues, each of which forms a discrete piece of the whole. The

first of these is the basic question of whether, and if so when,

an omission should be regarded as a statement of fact covered

under the common law of deceit, the federal securities laws,

and the Martin Act. The answer, as will become clear, is that

these omissions must be covered by the same contextual stan-

dards that are used with actual statements of fact. In this con-

text, the correct impulse is to expand the basic liability for de-

ceit, for there is no doubt that omissions can be as telling, or

material, as statements of fact. The second issue relates to

statements of opinion as distinct from statements of bald fact,

and the third relates to predictions of future events, a subset of

opinion. In both of these latter two contexts, it becomes criti-

cal to distinguish between “puffing,” i.e., vague commenda-

tions of goods and services, and specific statements that are

properly the source of liability in all three settings. I examine

each of these subcases after addressing the general conception

of materiality.

1. Materiality

The inquiry into statements of fact is framed by the legal

inquiry into “materiality.” Although Section 525 of the Second

Restatement of Torts does not use the term, it is widely under-

stood that any misrepresentation must involve some “material”

or major fact to be actionable. Otherwise any minor or incon-

sequential remark could generate a large liability, which is why

traditional writers on the topic spoke of actionable misrepresen-

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 831

tations, as a subclass of the larger whole.

The exact formula-

tion of materiality necessarily treats a question of degree as

though it were a hard line. Nonetheless, since speech is ubiq-

uitous in a way in which liability determinations cannot be,

there needs to be some initial filter that keeps a large amount

of small talk out of the legal system. How this is best done is

subject to much dispute. There are always ample grounds, for

example, to differ as to whether the determination of material-

ity turns on an independent judicial assessment of the impor-

tance of the statement, or whether materiality should be mea-

sured by reference to the specific persons to whom it is made,

which raises extra difficulties for any statement made to that

diverse group of individuals known as the public at large.

The definitive Supreme Court precedent on this matter is

TSC Industries, Inc. v. Northway,

which involved charges that a

proxy statement issued in connection with a proposed acquisi-

tion of TSC was “incomplete and materially misleading.”

addressing this question, the Supreme Court opted for the ob-

jective standard by explicitly rejecting the broader definition of

materiality adopted by the Seventh Circuit, which had used the

phrase “all facts which a reasonable shareholder might con-

sider important.”

The Supreme Court’s narrower standard

followed “the conventional tort test of materiality whether a

reasonable man would attach importance to the fact misrepre-

sented or omitted in determining his course of action.”

That

answer seems correct in the context of any public offering, for

it is not possible for any offeror to take into account the multi-

ple individual variations between different prospective sellers

(or buyers, as the case may be). But the subjective standard

could prove more acceptable in those cases where there is a

one-on-one transaction in which the defendant is aware of the

weakness and foibles of his trading partner. Clearly, for most

80. Hence the title of George Spencer Bower’s book, Actionable Misrepre-

sentation (1911), an early masterpiece on this topic. It is now in its fifth edi-

tion.

PENCER

OWER

& K

ANDLEY

, A

CTIONABLE

ISREPRESENTATION

(5th

ed. 2014).

81. TSC Indus., Inc. v. Northway, 426 U.S. 438 (1976).

82. Id. at 441.

83. Id. at 445 (quoting Northway, Inc. v. TSC Indus., Inc., 512 F.2d. 324,

330 (7th Cir. 1975)).

84. Id. (citing

ESTATEMENT

ECOND

)

ORTS

§ 538 (2)(a) (Tentative

Draft No. 10, 1964)).

832 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

purposes, the public setting dominates. In that context, the

Supreme Court in TSC Industries put the words “would” and

“might” in italics to stress the difference between these two def-

initions. The Court then cashed out this view by insisting that,

in the public context of investment decisions, “there must be a

substantial likelihood that the disclosure of the omitted fact

would have been viewed by the reasonable investor as having

significantly altered the ‘total mix’ of information made availa-

ble.”

Although sensible in its basic outlook, this principle in ap-

plication often overlooks some of the complexities of deciding

what kinds of statements or omissions should create potential

liability in dealing with acquisitions and mergers of publicly

traded stock. The best illustration of the hybridization of state-

ments and omissions is the 1988 Supreme Court decision in

Basic, Inc. v. Levinson,

where the question before the Su-

preme Court was whether it was a material misrepresentation

to say that a company’s board was not in preliminary merger

negotiations when in fact it was. Any release of this informa-

tion is obviously very ticklish, for its disclosure could push

stock prices upwards, which could make the deal less attractive

for the acquiring party.

But if mum is the word, what should

be done in response to a direct question about whether these

negotiations took place? The district court took the position

that these statements were not material because they were not

“destined” to lead to an agreement in principle.

As a matter

of simple economics, its broad contention must be wrong, be-

cause anyone thinking about buying shares in a company will

vary his estimate of their value, up or down, given knowledge

of the possible merger agreement, so that the price will move,

either up or down, on the information in the short run no

85. TSC Indus., 426 U.S. at 449.

86. Basic, Inc. v. Levinson, 485 U.S. 224 (1988).

87. The same risk arises when insiders, like lawyers or investment advi-

sors, trade on information supplied by the acquirer on their own account,

driving up the price of the stock. It is for this reason that the agent binds

itself by contract not to trade on that information. See United States v.

O’Hagan, 521 U.S. 642, 652–53 (1997) (applying the insider trading statute

to these misappropriation cases). For a critical analysis of this issue, see Rich-

ard A. Epstein, Returning to Common-law Principles of Insider Trading After New-

man v. United States, 125

ALE

L.J.

1482 (2016).

88. Basic, 485 U.S. at 229.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 833

matter what the ultimate outcome of the acquisition effort.

What is evident about this situation is that the reason for keep-

ing quiet is the same as the reason for making a misleading

statement. Here is one class of disclosures that hurts share-

holders as a group. So long, therefore, as management is not

trading on its own account, the statements it made should be

regarded as justified as the only effective means to preserve

the needed secrecy in the negotiations.

Unfortunately, securities law does not have a well-devel-

oped category of justified falsehoods, so the hard question be-

comes how to determine the materiality of various noncom-

mittal board responses over the life cycle of merger and acqui-

sitions negotiations, many of which fail. The laundry list of

factors offered in Basic

is so comprehensive that it is beyond

useless, so that in practice the only way for firms to handle the

question in practice is to adopt a blanket approach by which

they always say “no comment” whenever asked about possible

merger negotiations. That sphinxlike response conveys no in-

formation, but it is far from clear whether that private adapta-

tion is more efficient than a legal regime that does allow for

justified concealment. More specifically, there may well be

cases in which it makes sense for the firm to let the market

know that there is no pending merger so that they can better

price the shares. But that option is precluded by Basic.

To make matters more difficult, the proof of reliance by

individual traders—the causal link that will be discussed in de-

tail later

—does not mesh well with the facts in Basic that led

89. Id. at 239, which states that:

Whether merger discussions in any particular case are material

therefore depends on the facts. Generally, in order to assess the

probability that the event will occur, a factfinder will need to look

to indicia of interest in the transaction at the highest corporate

levels. Without attempting to catalog all such possible factors, we

note by way of example that board resolutions, instructions to in-

vestment bankers, and actual negotiations between principals or

their intermediaries may serve as indicia of interest. To assess the

magnitude of the transaction to the issuer of the securities alleg-

edly manipulated, a factfinder will need to consider such facts as

the size of the two corporate entities and of the potential premiums

over market value. No particular event or factor short of closing the

transaction need be either necessary or sufficient by itself to render

merger discussions material.

90. See infra Section II.E.

834 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

to the fraud-on-the-market theory and posits that the price of

shares in the market reflects all “available material informa-

tion regarding the company and its business,” even if there are

no acts of individual reliance.

Hence the fraud-on-the-mar-

ket theory is introduced to fill that gap so as to provide a

rough estimate of the price differential at the time of the false

announcement. But the situation is ultimately unstable be-

cause those who sold improvidently recouped their losses,

while those who purchased fortuitously get to keep their gains,

so that shareholders at the time of the judgment suffer the

diminution in value. It is no wonder that the all-silent ap-

proach dominates any potential remedy given, proving once

again the difficulty in using these theories to establish liability.

As will become evident, theories of liability that work reasona-

bly well in direct negotiations do not always work as well in

connection with public offerings with the diffuse distribution

of information.

The complexities on the topic of materiality at common

law, under the federal securities laws, and under the Martin

Act, are compounded when the plaintiff’s case involves not a

single statement but an elaborate course of dealing in which

multiple statements are made in multiple arenas. More specifi-

cally, many financial fraud cases do not result from a single

statement by one party to another. Instead there is a constant

stream of information that may be accurate in part, but inaccu-

rate in some respects and incomplete in others. Any effort to

control fraud that looks at the former but ignores the latter is

woefully incomplete, so that any effective control of fraud

yokes together the treatment of omissions and misrepresenta-

tions.

That interplay between material statements and omissions

was very much in play in the important Supreme Court deci-

sion Matrixx Initiatives, Inc. v. Siracusano,

a spiritual sequel to

TSC Industries. Matrixx was the defendant in a class action for

securities fraud brought under Section 10(b) of the Exchange

Act of 1934. The gist of the complaint was that the defendant

had made a material omission when it had failed to disclose

that it had received information from reliable physicians who

91. Basic, 485 U.S. at 241–42 (quoting Peil v. Speiser, 806 F.2d 1154,

1160–61 (3d Cir. 1986)).

92. Matrixx Initiatives, Inc. v. Siracusano, 563 U.S. 27 (2011).

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 835

had detected a high level of anosmia, or loss of smell, among

individuals who had used its main product, Zicam, a cold rem-

edy. The previous controlled studies on the topic had not, as

Matrixx had stressed, revealed any systematic connection be-

tween the two, even though other studies had noted that anos-

mia had occurred in individuals who had used other zinc-

based products. The defendant’s contention was that the com-

plaint did not adequately allege that Matrixx made a material

representation or omission or that it acted with scienter be-

cause the complaint did not allege that Matrixx “knew of a

statistically significant number of adverse events requiring dis-

closure.” A unanimous Supreme Court held that it was proper

to let the case go to the jury because any analysis of “the mate-

riality of adverse event reports cannot be reduced to a bright-

line rule.”

Ironically, the Court’s analysis seems wrong on both

points. The evidence in this case did support a bright-line re-

sult, but for the plaintiff, not the defendant. While there are

often situations for which bright line rules are hard to man-

age, this case was not one of them. Adverse events are always a

big deal in medicine, wholly apart from their financial implica-

tions.

A single adverse event might give rise to a jury ques-

tion of whether it is a powerful enough signal to warrant a

directed verdict for the plaintiff. But usually reports of adverse

events come in bunches, and the case is certainly not a jury

question when an extended series of documented cases reveal

the same adverse consequence, whose seriousness is not in

doubt.

To give two points of reference in this case, it is clear be-

yond any doubt that a plaintiff in a medical-malpractice case

would have slam-dunk proof that the omission of this informa-

tion created a breach of the physician’s duty to disclose possi-

ble material complications from the proposed treatment.

Worse still, as in Matrixx, the FDA had received multiple

clinical reports on the side effects of the drug, on the strength

93. Id. at 30.

94. For a glimpse of how big the impact can be, see

FDA A

DVERSE

VENTS

EPORTING

YSTEM

(FAERS) P

UBLIC

ASHBOARD

, https://www.fda.gov/drugs

/guidancecomplianceregulatoryinformation/surveillance/adversedrugeffec

ts/ucm070093.htm.

95. For the basic doctrine, see Canterbury v. Spence, 464 F.2d 772, 783

(D.C. Cir. 1972).

836 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

of which it launched its own investigation. Days later, the de-

fendants issued a press release stating that the FDA had not

found any connection between Zicam and anosmia, which

knowingly gives a false impression about the impact that an

FDA study would have on the marketability of the drug.

this situation, moreover, the omission in question was mani-

festly material because the mere fact that drug is subject to

FDA investigation is itself sufficient to drive down the share

price even if the drug is ultimately exonerated. In cases where

there are these obvious benchmarks keyed to the notion of

materiality, there is no need for a jury determination. As is so

often the case, the common strategy of courts in tort cases is to

allow juries to decide cases when a cleaner appreciation of the

underlying substantive law supports a partial summary judg-

ment for the plaintiff.

2. Omissions

In light of the fact pattern in Matrixx and similar cases, it

should come as no surprise that even though the term “omis-

sion” is not found in the Martin Act, the term has, by judicial

decision, always been read into the definition.

Hence the

overall analysis takes place in the same fashion as it does when

the term “omission” is included, as in Section 11 of the Securi-

ties Act dealing with registration statements.

Section 11 cov-

ers strict liability in a registration statement that “contained an

untrue statement of a material fact or omitted to state a mate-

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

statements therein not misleading.”

Nonetheless, in practice,

the actual language of the governing law takes a back seat to a

functional analysis in ways that expand the operative scope of

the Martin Act. Any isolated utterance resists classification un-

96. See Matrixx, 563 U.S. at 34–35.

97. See, e.g., People ex rel. Vacco v. World Interactive Gaming Corp., 714

N.Y.S.2d 844, 864 (Sup. Ct. 1999) (“It is well settled that fraud exists not only

where there has been an affirmative misstatement of a material fact, but also

where there has been an omission of a material fact.”).

98. 15 U.S.C. § 77k(a) (“In case any part of the registration statement,

when such part became effective, contained an untrue statement of a mate-

rial fact or omitted to state a material fact required to be stated therein or

necessary to make the statements therein not misleading, any person acquir-

ing such security . . . may . . . sue.”).

99. Id.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 837

til the background context, whether specific to an individual

case or socially generalizable, is understood. As in Matrixx, to

say a good fact about X—the FDA had not found any negative

correlation between Zicam and anosmia—only to omit a bad

fact about X—the FDA was investigating recent reports of that

connection—is to speak a half-truth, which is a lie if the omis-

sion is deliberate and is all the more insidious because the

half-truth lends an aura of plausibility to the overall transac-

tion. To turn a blind eye to these omissions would allow fraud-

sters to disarm their listeners en masse with impunity.

Doctrinally, therefore, it is no surprise that the Martin Act

has always been construed in ways that bring it in line with the

federal securities laws and the common law on the general is-

sue of materiality and the specific problem of omissions.

100

In-

deed, in New York v. Rachmani Corp.,

101

the New York Court of

Appeals explicitly adopted the definition of materiality articu-

lated in TSC Industries, and further noted that “[a]n omitted

fact is material if there is a substantial likelihood that a reasona-

ble shareholder would consider it important in deciding how

to vote.”

102

The statement was made in connection with indi-

viduals who voted in favor of a cooperative sale when they

lacked full information about the transaction. As with all omis-

sions, there is no obvious way to rely on what has not been

said.

103

The use of the term of “omitted fact” was essential to

Rachmani because of the nature of the underlying dispute,

namely the conversion of rental units to condominiums. The

problem arose because of “the failure to mention an unsatis-

fied precondition to a cooperative conversion of an apartment

house constituted fraud which justified the issuance of an in-

junction” under the Martin Act.

104

The New York Court of Ap-

peals denied injunctive relief on the ground that there was no

“actionable” fraud because the disputed condition was not sig-

100. For the industry view, see

ECHERT

LLP, T

ARTIN

: A P

RIMER

(Jan. 15, 2004), https://www.dechert.com/content/dam/dechert%20files/

onpoint/2004/1/new-york-state-s-martin-act-a-primer/FS_2004-04.pdf.

101. New York v. Rachmani Corp., 525 N.E.2d 704, 726–27 (N.Y. 1988).

102. Id. at 726.

103. See Affiliated Ute Citizens v. United States, 406 U.S. 128, 153 (1972).

104. Rachmani Corp., 71 N.Y.2d at 721.

838 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

nificant and had, in any event, been disclosed in the original

offering plan.

105

The conceptual point, however, is that in all cases, the

tort of deceit cannot be limited to false statements, but must

also take into account omissions that when set against the

background of what was discussed and mentioned, could have

an impact on the course of conduct in a particular case.

106

But

here too there is nothing in the architecture of the Martin Act

which makes the common law or federal securities law analysis

on this critical initial coverage issue obsolete, otiose, or irrele-

vant.

3. Statements of Fact and Opinion

The basic inquiry follows similar lines in drawing the dis-

tinction between statement of facts and statement of opinions.

It is easy enough to conjure statements of the latter. The radio

commercial on WFMT that states that the Goodman Theatre is

the “foremost” theater in Chicago may sound like a statement

of fact, but is, in fact, widely regarded as a statement of opin-

ion, because of the drastic consequences that would flow from

the opposite interpretation. There are many theaters in Chi-

cago, and there are surely some rankings that hold that the

Goodman is not the best theater, or perhaps even in the top

ten. The same analysis applies to heaping praise on athletes,

entertainers, CEOs, or just about anyone else. There is a

strong and justified sense that the public at large, especially

the universe of theater-goers who care about these matters, is

context-sensitive, and will appropriately discount any claim of

supposed preeminence that is made in such bald terms.

One common way to put the point is to note that the

ubiquitous practice of “puffing” is not actionable even if liter-

ally false. One standard account of what falls into a safe haven

is offered by Learned Hand in Vulcan Metals Co. v. Simmons

Mfg. Co.

107

105. Id.

106. It is common for a trial judge to instruct the that “the definition of a

‘false statement’ included both affirmative misrepresentations and any rep-

resentation that ‘effectively conceals or omits a material fact.’” See People v.

Sala, 739 N.E.2d 727, 730 (N.Y. 2000).

107. Vulcan Metals Co. v. Simmons Mfg. Co., 248 F. 853 (2d Cir. 1918).

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 839

There are some kinds of talk which no sensible man

takes seriously, and if he does he suffers from his cre-

dulity. If we were all scrupulously honest, it would not

be so; but, as it is, neither party usually believes what

the seller says about his own opinions, and each

knows it.

108

A social code, universally accepted, vanquishes any hyper-

bolic literalism in thinking about ordinary discourse. Until the

claims in question become more specific, the better approach

by far is to let matters slide, so that self-correction takes place

on the side of the listener, in light of public facts. But when

the nature of the claim is specific, such as “Good Housekeeping

has ranked our product as best in class,” it is a false statement

of fact if Good Housekeeping has done no such thing. The inevi-

table hard cases therefore arise when the implicit background

understandings do not hold, precisely because one side has a

hidden store of information that the other lacks.

One early case that shows a deep appreciation of this dis-

tinction and all it entails is Smith v. Land and House Property

Corp.,

109

in which the plaintiff had offered a hotel for sale, stat-

ing that it was then let to “‘Mr. Frederick Fleck (a most desira-

ble tenant).’”

110

Before the contract of sale could be signed,

Fleck made several late payments on the lease and then went

into bankruptcy. When the defendant refused to complete the

transaction on the ground that the plaintiff’s statement about

Fleck was a misrepresentation, the plaintiff sought specific per-

formance, claiming that their gauzy remark was “a mere ex-

pression of opinion.”

111

Lord Justice Bowen rejected that con-

tention, hitting all the right notes:

If the facts are not equally known to both sides, then

a statement of opinion by the one who knows the

facts best involves very often a statement of a material

fact, for he impliedly states that he knows facts which

justify his opinion. Now a landlord knows the rela-

tions between himself and his tenant, other persons

either do not know them at all or do not know them

equally well, and if the landlord says that he consid-

108. Id. at 856.

109. Smith v. Land & House Prop. Corp. (1884) LR 28 Ch D 7 (Eng.).

110. Id. at 8.

111. Id. at 10.

840 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

ers that the relations between himself and his tenant

are satisfactory, he really avers that the facts pecu-

liarly within his knowledge are such as to render that

opinion reasonable. Now are the statements here

statements which involve such a representation of

material facts? They are statements on a subject as to

which prima facie the vendors know everything and

the purchasers nothing. The vendors state that the

property is let to a most desirable tenant, what does

that mean? I agree that it is not a guarantee that the

tenant will go on paying his rent, but it is to my mind

a guarantee of a different sort, and amounts at least

to an assertion that nothing has occurred in the rela-

tions between the landlords and the tenant which

can be considered to make the tenant an unsatisfac-

tory one. That is an assertion of a specific fact. . . . In

my opinion a tenant who had paid his last quarter’s

rent by driblets under pressure must be regarded as

an undesirable tenant.

112

The fact/opinion distinction plays out the same way in

defamation cases as in deceit cases. Thus, suppose some third

party had made the same statement that Frederick Fleck was a

most desirable tenant to another third party, who then pur-

chased the property from his landlord who had said nothing,

or that a third party had deliberately understated the size of

the farm that the plaintiff purchased, likewise when the owner

had said nothing. In both settings, the plaintiff took the risk as

against the owner, but not against the third party, because a

defendant in a defamation case could not defend on the

ground that the basic statement was simply a matter of opinion

if he knew as much as the defendant did in Smith. The still

classic 1910 treatment of the issue by Van Vechten Veeder

took just this view.

113

Let the underlying facts be stated before

112. Id. at 15–16.

113. See Van Vechten Veeder, Freedom of Public Discussion, 23

ARV

. L. R

413, 419–20 (1910):

The distinction is fundamental, then, between comment upon

given facts and the direct assertion of facts. And the significance of

the distinction is plain. If the facts are stated separately, and the

comment appears as an inference drawn from those facts, any injus-

tice that the imputation might occasion is practically negatived by

reason of the fact that the reader has before him the grounds upon

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 841

the conclusion is reached, then third parties can make their

own judgment whether the evidence presented supports the

conclusion. Hence a detailed account of a transaction that is

then branded a theft is not defamatory if the underlying fac-

tual statements are true. But if the charge of theft is made

without that supporting context, the identical statement is now

transformed from one of opinion based on known facts to a

statement of fact that can support liability for either defama-

tion or deceit. The one obvious exception is where the un-

stated facts are so widely known to the public from multiple

sources that there is no reason to repeat them each time the

topic arises, because the requisite symmetry of which Lord Jus-

tice Bowen spoke is again present. The bottom line is that it is

the presence of asymmetrical information that controls the

characterization of any given statement as one of fact or opin-

ion.

This same limitation is carried over into securities law

where again the question is whether a given statement escapes

sanction because it counts as one of opinion, on the tests out-

lined above. This issue came to the fore in the recent Supreme

Court decision in Omnicare, Inc. v. Laborers District Council Con-

struction Industry Pension Fund,

114

which is notable for how Jus-

tice Elena Kagan first relied on the Restatement (Second) of

Torts § 539,

115

then on a brief excerpt from Bowen’s opinion

which the unfavorable inference is based. When the facts are truth-

fully stated, comment thereon, if unjust, will fall harmless, for the

former furnish a ready antidote for the latter. The reader is then in

a position to judge whether the critic has not by his unfairness or

prejudice libeled himself rather than the object of his animadver-

sion. But if a bare statement is made in terms of a fact, or if facts

and comment are so intermingled that it is not clear what purports

to be inference and what is claimed to be fact, the reader will natu-

rally assume that the injurious statements are based upon adequate

grounds known to the writer. In one case, the insufficiency of the

facts to support the inference will lead fair-minded men to reject it;

in the other, there is little, if any, room for the supposition that the

injurious statement is other than a direct change of the fact, based

upon grounds known to the writer, although not disclosed by him.

114. Omnicare, Inc. v. Laborers Dist. Council Constr. Indus. Pension

Fund, 135 S. Ct. 1318 (2015).

115.

ESTATEMENT

ECOND

)

ORTS

§ 539 (

. L

NST

1977).

(1) A statement of opinion as to facts not disclosed and not other-

wise known to the recipient may, if it is reasonable to do so, be

interpreted by him as an implied statement

842 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

in Smith, to the effect that when “the facts are not equally

known to both sides, then a statement of opinion by the one

who knows the facts best . . . impliedly states that [the speaker]

knows facts which justify his opinion.”

116

Omnicare arose under Section 11 of the 1933 Securities

Act, which, as mentioned previously, subjects the registration

statement of an issuer to liability if that statement “contained

an untrue statement of a material fact or omitted to state a

material fact required to be stated therein or necessary to

make the statements therein not misleading.”

117

In connection

with a new public offering of common stock, Omnicare stated:

“We believe our contract arrangements with other healthcare

providers, our pharmaceutical suppliers and our pharmacy

practices are in compliance with applicable federal and state

laws.”

118

It further averred: “We believe that our contracts with

pharmaceutical manufacturers are legally and economically

valid arrangements that bring value to the healthcare system

and the patients that we serve.”

119

Both statements were ac-

companied by caveats; Omnicare “mentioned several state-ini-

tiated ‘enforcement actions against pharmaceutical manufac-

turers’ for offering payments to pharmacies that dispensed

their products.”

120

It then cautioned that the laws relating to

that practice might “be interpreted in the future in a manner

inconsistent with our interpretation and application.”

121

The plaintiffs sued for violation of the strict-liability Sec-

tion 11, claiming that defendant’s representations were “mate-

rially false” because the federal government later investigated

Omnicare for violation of its anti-kickback laws.

122

The pattern

necessarily raises the much-mooted interconnection between

statements of fact and opinion. Justice Kagan thought that the

(a) that the facts known to the maker are not incompatible with

his opinion; or

(b) that he knows facts sufficient to justify him in forming it.

(2) In determining whether a statement of opinion may reasonably

be so interpreted, the recipient’s belief as to whether the maker has

an adverse interest is important.

116. Omnicare, 135 S. Ct. at 1330–31.

117. 15 U.S.C. § 77k(a) (1933).

118. Omnicare, 135 S. Ct. at 1323.

119. Id.

120. Id.

121. Id.

122. Id. at 1324.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 843

last statement could be actionable given that it did not men-

tion the possibility of the federal investigation, but only state-

initiated actions. In my view, the gap between this statement

and one treating Frederick Fleck as a desirable tenant when

known to be in arrears on his rent and teetering on bank-

ruptcy seems enormous because of the differences in the back-

ground knowledge of the two target audiences. The Omnicare

investors, often represented by learned intermediaries, were

not in the dark about the general structure of the securities

laws and the extensive substantive overlap between state and

federal laws regulating every aspect of the health care business.

Nor did they lack the common knowledge that government

investigations of billing standards and practices are a fixed and

inescapable part of any ongoing health-care business. There-

fore, it seems hard to believe that anyone who is apprised of

the risk of investigation at the state level can somehow be bliss-

fully ignorant of the parallel risk at the federal level, knowing

that overlapping federal and state enforcement systems is the

norm in virtually every regulatory area today.

It is possible to go further. Whenever a dispute arises

about regulatory risk, the close case is one in which the pro-

spectus says nothing about regulatory risk at any level. Even

here, there is much to be said for the view that everyone knows

that healthcare is a heavily regulated industry, so that the risk

of an administrative investigation is too obvious to belabor.

But even if, as seems the case, that somewhat jaded view is in-

correct, the level of inference rightly expected from sophisti-

cated buyers is so powerful that it becomes hard, even if this

supposed omission is given weight, to credit the claim for its

materiality in light of all the obvious inferences that can be

drawn from the extensive set of stated and background infor-

mation. And there is a heavy cost for flyspecking these registra-

tion statements, for the volume of offerings will diminish sub-

stantially if virtually every registration can be found with the

benefit of hindsight to contain some material exaggeration or

omission, under a strict liability regime no less. The oft-ex-

pressed fear of common-law judges was that expanding the

class of false statements of fact could by degrees make every

speaker a warrantor of the success of any deal that turns sour.

Indeed, Justice Holmes said it well: “The rule of law [that

“vague commendation” of wares is not actionable] is hardly to

be regretted, when it is considered how easily and insensibly

844 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

words of hope or expectation are converted by an interested

memory into statements of quality and value, when the expec-

tation has been disappointed.”

123

In my view, a summary judg-

ment for the defendant in Omnicare puts a clear limit on the

scope of misrepresentations of fact that lines up well with the

common-law theory that should animate this entire area.

4. Predictions

The same conceptual framework carries over to the line

between statements of fact, which are actionable under the se-

curities law, and “mere” predictions, which are not. Often the

initial cut into the problem stems from the quotation from

Learned Hand in Vulcan about the line between mere puff

and an actionable misrepresentation.

124

The distinction reso-

nates in this context, because predictions deal with the future,

which is often highly uncertain, and not the past, whose events

are fixed even if not always known with precision. In this con-

text, it is well-nigh universally held that statements of “corpo-

rate optimism” of the sort which insist that sales are “going

reasonably well,” or that the company was “pursuing the right

strategy,” or that the firm will “start to grow revenue” in the

second half of the year, are regarded as non-actionable puffery

by modern judges,

125

and for the identical reason expressed by

Lord Justice Bowen and Judge Hand:

[N]ot every unfulfilled expression of corporate opti-

mism, even if characterized as misstatement, can give

rise to a genuine issue of materiality under the securi-

ties laws. . . . In particular, courts have demonstrated

a willingness to find immaterial as a matter of law a

certain kind of rosy affirmation commonly heard

from corporate managers and numbingly familiar to

the marketplace—loosely optimistic statements that

are so vague, so lacking in specificity, or so clearly

constituting the opinions of the speaker, that no rea-

123. Deming v. Darling, 20 N.E. 107, 108–09 (N.Y. 1889).

124. See N.Y.C.

DMIN

. C

ODE

§§ 26-501 to 26-520, and supra text accompa-

nying note 75.

125. Shaw v. Dig. Equip. Corp., 82 F.3d 1194, 1217–19 (1st Cir. 1996),

abrogated on other grounds by 15 U.S.C. § 78u-4(b)(2) (1934).

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 845

sonable investor could find them important to the to-

tal mix of information available.

126

Judicial pronouncements of this sort in the decided cases

can be multiplied at will,

127

and the degree of unanimity of

opinion over the question covers not only federal securities

laws but also state securities laws.

128

The logic is again the

same. It is not enough to look only at the statement as made; it

is also necessary to take into account the general background

knowledge of the audience, whose members are unlikely to be

taken in by bland, commonplace predictions of future com-

mercial success. All of these observations on facts, opinions,

and predictions carry over to the Martin Act, where exactly the

same dynamic takes place.

The moral is that on key points of doctrine, the differ-

ences between the common law, the federal law, and the Mar-

tin Act are overstated. It is easy to miss this point if one fails to

appreciate that the common law is not just some arbitrary as-

semblage of rules but instead a coherent intellectual structure

for which it is difficult (especially on short notice) to gin up a

rival. Accordingly, the best cases from over a century ago sur-

vive undiminished today. Therefore, if it is impossible to en-

join these statements under the Martin Act, it would become

indefensible to think that such statements should be the

source of a damage suit, let alone a criminal prosecution. So

much of the interpretation of the securities laws is sensible

gloss that the lessons learned from common-law cases and fed-

eral securities laws should not be ignored in dealing with the

Martin Act, and for the most part, judicial decisions respect

the overlap in analysis on the points of commonality.

E. Causation

A good deal of attention must also be paid, moreover, to

the causal issues implicit in any case of fraud, or even misrepre-

sentation. As a matter of basic theory, causation is a key ele-

126. Id. at 1217.

127. See, e.g., Robert N. Kravitz

Room for Optimism: The “Puffery” Defense

under the Federal Securities Laws (Part 1 of 2), 19 A.B.A.

. L

ITIG

. J.,

Winter

2009, at 23; Kravitz

Room for Optimism: The “Puffery” Defense under the Federal

Securities Laws (Part 2 of 2), 19 A.B.A.

. L

ITIG

. J.

Spring 2009, at 1. See also

ECA v. J.P. Morgan Chase & Co., 553 F.3d 187 (2d Cir. 2009).

128. See, e.g., Next Century Commc’ns Corp. v. Ellis, 318 F.3d 1023 (11th

Cir. 2003) (applying Georgia law).

846 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

ment in all torts cases, whether for physical injury or financial

losses, because it links up the act or omission of the defendant

to the plight of the plaintiff. The simple assertion that the

plaintiff has been hurt is consistent with the harm being self-

inflicted, being the result of an act of God, or being inflicted

by any third party unrelated to the defendant. To be sure, no-

fault systems compensate injured persons regardless of the

cause of injury. However, under that approach, the compensa-

tion always comes from a general fund raised by tax revenues

or some insurance scheme like employer-based workers’ com-

pensation—not from a specific exaction collected from one

isolated person among millions who is singled out for special

treatment.

The same logic that applies to damages awards also ap-

plies to injunctive relief. Any particular person is entitled to an

injunction only if she faces the prospect of some imminent

harm, including the repetition of some past adverse event. But

that injunction can only be directed to the person who caused

the harm or created the risk. When these harms are global,

either a class action or some direct government enforcement

is appropriate, under the rules of public nuisance that have

been well established for nearly 500 years.

129

The basic line is

that special harms to a single person are remedied by a direct

action, while those low-level harms that are broadly felt are

remedied by an administrative action by some public body

against the wrongdoer that could both fix the wrong and im-

pose a fine for its occurrence. (In the latter case, the funds in

question stay in the public treasury, typically to be spent on,

say, the road or waterway that has been subject to damage.)

The same general logic can apply to false public statements or

omissions in which it is impossible to pinpoint the individuals

who have suffered the damages. These basic rules apply to

omissions as well as acts of commission, and the usual, and

correct, approach is to first find some statutory or private law

duty that requires some action, and then hold the failure to

discharge that duty an actionable wrong. In some cases, these

duties are strict, but in most they tend to require some level of

care that can vary by circumstances. But causation remains an

element in both kinds of cases.

129. Y.B. Mich. 27 Hen. 8, f. 27, pl. 10 (1536) (Eng.).

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 847

In dealing with physical harms, respecting the rules on

remoteness of damage for both acts and omissions are essen-

tial to the overall integrity of the system. To be sure, the for-

mal test of causation in the law of torts relies on an unhappy

amalgam of “but for” causation, limited by some notion of

proximate causation, based alternatively on foresight or direct-

ness.

130

But this two-part test makes matters more complicated

than they really are, for in practice, most cases of causation

work backwards from the injury to a particular defendant, and

not forward from a distant act of some potential defendant to

the injury of the plaintiff.

131

Thus, the paradigmatic case of

causation is the direct application of force—hands, sticks,

guns, and the like—covered by the common law rules of tres-

pass. The complex variations on it are those cases of indirect

harm where the defendant sets a dangerous condition that re-

sults in harm to the plaintiff by the application of some inter-

mediate force.

132

In these situations, the damages are too re-

mote for liability when the dangerous condition created by the

defendant is corrected, as when a trap is filled before anyone

can fall in it, or a bomb is defused before it explodes. If some

other person resets the trap or rearms the bomb, responsibility

for the harm should lie with the later actor. Only in very rare

cases could one go back behind that action to the remote

stranger on an expanded theory of causation.

A parallel form of analysis applies to the tort of misrepre-

sentation at common law, and by extension to all actions that

take place under the federal securities laws or the Martin Act

and other blue-sky laws. In this context, the key element, as

articulated in all the cases discussed above, is that of the plain-

tiff’s reliance on the misstatement of the defendant. Omissions

also are generally judged by a test that asks whether the provi-

sion of additional information would have altered the plain-

tiff’s conduct in whole or in part in a way that would have re-

130. See

ESTATEMENT

HIRD

)

ORTS

: L

IABILITY FOR

HYSICAL AND

MO-

TIONAL

ARM

§ 26 (

. L

NST

. 2010)

131. See Richard A. Epstein, A Theory of Strict Liability, 2

J. L

EGAL

TUD

151

(1973);

ICHARD

A. E

PSTEIN

& C

ATHERINE

M. S

HARKEY

, C

ASES AND

ATERIALS

ORTS

341–43 (11th ed. 2016).

132. For the classic discussion, see Scott v. Shepherd (1773) 96 Eng. Rep.

525 (KB). In Roman law, the same distinction applies between killing (oc-

cidere) and furnishing a cause of death (causam mortis praestare).

. 9.2.7.1

(Ulpian, Ad Edictum 18);

. 9.2.51.pr.

848 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

duced or avoided the loss. There is no tighter test, and giving

up on causation in omission cases is tantamount to a huge

compromise of the entire system of public and private enforce-

ment. Where that causal connection is established in particu-

lar cases, an award of damages is appropriate for harm suf-

fered. Where it is threatened, or even imminent, in connec-

tion with speech directed to a specific person, private

injunctions are not likely to play much of a role because any

person who is smart enough to request a private injunction

against being deceived or misled is smart enough not to need

it. Armed with the truth, he should never be deceived by a

falsehood. But public injunctions, such as those contemplated

under both the federal securities laws and the Martin Act, re-

main fitting because it is perfectly appropriate to protect un-

suspecting members of the public from systematic frauds which,

if allowed to go forward, would result in substantial financial

losses. Fraud on the public is thus an obvious analogue to the

common-law public nuisance.

The role of reliance in these cases is made clear by this

simple, but often overlooked, distinction. It is one thing to lie

to another person, but it is quite another to deceive that person.

In the former case, the target of the lie is not misled, or even if

misled, did not take any steps that led to his own damage. The

lie is an attempt at deceit, but, like the blow that does not land

or the bomb that does not go off, it is not a completed tort.

The point is evident from the grammar of both verbs. “I lied to

you” is not an equivalent to “I deceived you.” Reliance by the

particular plaintiff on the false statement of the defendant is

thus needed to complete the private wrong. As Judge Francis

Buller said in the landmark 1789 case of Pasley v. Freeman,

133

“Fraud without damage, or damage without fraud, gives no

cause of action; but where these two concur, an action lies.”

134

In Pasley, the defendant stated to the plaintiff that Falch

was worthy of credit, but the defendant knew that he was in

fact not creditworthy. The novel feature of the case was that

the fraud in question appeared to be not for the plaintiff’s di-

rect benefit, so that one judge, Judge Grose, took the position

that no liability should attach because there was no collusion

133. Pasley v. Freeman (1789) 100 Eng. Rep. 450 (KB). The decision to

allow the action was 3 to 1.

134. Id. at 453.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 849

between the defendant and Falch and there was no contrac-

tual privity between the plaintiff and the defendant, given that

the loan was made to a third person.

135

But the three judges in

the majority were not troubled by the want of a return benefit.

Indeed, they regarded the defendant’s action as being “more

diabolical” because of the lack of an immediate self-interest.

136

But for present purposes, the key element is that the plaintiff

relied explicitly and immediately on the defendant’s state-

ments about the creditworthiness of Falch. In causal terms,

there was no possible intermediate action that could sever the

connection in this discrete financial context. It was not as

though the defendant had made this statement to some stran-

ger who in turn had relayed the information to the plaintiff.

That immediacy requirement is abundantly satisfied in cases

like Smith, with the face-to-face interactions between buyer and

seller.

137

F. Privity

The discussion of causation in Pasley opens yet another

complication in these situations, for the prospect of third-per-

son interventions raises serious questions as to whether the de-

faults suffered should be attributable to the defendant. In-

deed, the reference to privity obviously brings forward the

early history of the actions for harms caused by defective prod-

ucts after they left the hands of the defendant and passed

through intermediate hands. Thus, the early case of Winterbot-

tom v. Wright

138

held that the driver of a defective coach could

not recover from the defendant, who had supplied coaches to

the Postmaster General, who in turned leased them to the

plaintiff’s employer, when the plaintiff was injured when the

coach broke down because of a latent defect. The want of priv-

ity was held to defeat the action against the defendant who

had supplied and maintained the coach, even though the

plaintiff did not have an action against his employer unless he

could prove, which seemed unlikely, the latter’s improper

maintenance or use of it.

135. Id. at 451–53.

136. Id. at 456.

137. See supra notes 109–112 and accompanying text.

138. Winterbottom v. Wright (1842) 152 Eng. Rep. 402.

850 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

This legal doctrine slowly reversed itself through case af-

ter case in which latent defects caused damage to innocent

plaintiffs after going through the hands of one or more third

parties, such as distributors or retailers.

139

The question was

whether the use of that conduit, which severed privity of con-

tract, also made any damage suffered by the plaintiff too re-

mote. That rule applied to the full range of products, includ-

ing the poison belladonna sold as a harmless dandelion ex-

tract in Thomas v. Winchester,

140

the broken wheel in

MacPherson v. Buick Motor Co.,

141

or the alleged defective bottle

of Coca-Cola in Escola v. Coca-Cola Bottling Co.

142

In each of

these cases three conditions combined to limit liability. The

defect in question was latent; it reached the plaintiff in its orig-

inal condition; and the plaintiff made normal and proper use

of the product.

143

The mischief of the privity requirement—its

excessive restrictiveness—was well captured by Judge Cardozo

when he said: “The dealer was indeed the one person of whom

it might be said with some approach to certainty that by him

the car would not be used. Yet the defendant would have us

say that he was the one person whom it was under a legal duty

to protect. The law does not lead us to so inconsequent a con-

clusion.”

144

Yet if the traditional privity rule was too restrictive, its

abandonment opened the door too wide to liability, as all of

the limitations initially built into the early formulations of

product-liability law were consciously peeled away. Thus,

under the new regime that began to emerge in the 1960s, up-

stream manufacturers were systematically held liable for losses

from patent defects, downstream product modifications, or

plain misuse, all of which are better controlled either by inter-

mediate parties or by the plaintiff herself.

145

These shifts may

139. For my account of this early history, see

ICHARD

A. E

PSTEIN

ODERN

RODUCTS

IABILITY

–

(1980).

140. Thomas v. Winchester, 6 N.Y. 397 (1852).

141. MacPherson v. Buick Motor Co., 111 N.E. 1050 (N.Y. 1916).

142. Escola v. Coca-Cola Bottling Co., 150 P.2d 436, 468 (Cal. 1944) (Tray-

nor, J., concurring) (“The manufacturer’s liability should, of course, be de-

fined in terms of the safety of the product in normal and proper use, and

should not extend to injuries that cannot be traced to the product as it

reached the market.”).

143.

PSTEIN

, supra note 139.

144. MacPherson, 111 N.E. at 1053, quoted in Escola, 150 P.2d at 465.

145.

PSTEIN

supra note 139, at 76–92.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 851

seem small in the abstract, but they had huge practical conse-

quences. The insurance needed to cover losses under the Es-

cola strict-liability regime was a tiny fraction—less than one

percent—of what the new regime demanded.

146

Properly understood, the privity requirement served as a

rough but clearly imperfect proxy for the causation require-

ment. But in dealing with financial losses in all sorts of con-

texts, the privity requirement retains a much more tenacious

role. Information moves so quickly among persons that there

remains to this day a genuine fear that liability can easily go

too far. To put the point in its simplest form: no matter how

many times a single product is transferred, it remains a single

product—one drug tablet, one Coke bottle, one car. But every

time information is transferred, there is a huge multiplicative

effect. That information is retained by the original party and is

now shared, time and time again, with a new round of individ-

uals. In some of these iterations the information retains its

original form, but in many it is mixed and combined with in-

formation from other sources, so that it is hard to know exactly

how much of the original information remains.

These difficulties were apparent to Justice Cardozo in the

most influential case dealing with financial losses, Ultramares

Corp. v. Touche,

147

which he wrote fifteen years after his deci-

sion in MacPherson and which reveals a far different mindset

toward the privity issue. The case involved the movement of

information among three parties, in contrast to more public

offerings or statements. The defendant accountant failed to

detect the fraud of its client, Fred Stern & Co., which in turn

was able to procure extensive loans, both secured and un-

secured, from the plaintiffs. The complication arose because

Stern subsequently went bankrupt. In a first-best world, the ac-

146. In August 1976, I was hired by the American Insurance Association

(AIA), consisting of the major stock companies, to analyze the industry’s

large losses and premium hikes. At that time, the AIA had no systematic

institutional understanding of the issue. Most people thought that the strict

liability rule of the Restatement (Second) of Torts, Section 402A, was the

major cause of the change, but that was in fact wrong. It was the three ele-

ments just mentioned—expanded definition of defect, no insulation from

product modifications, and a broad foreseeable misuse standard—that drove

the analysis. For my then-contemporary account of the problem, see Richard

A. Epstein, Products Liability: The Search for the Middle Ground, 56 N.C. L.

643 (1978), which summarized the results of my work for the AIA.

147. Ultramares Corp. v. Touche, 174 N.E. 441 (N.Y. 1931).

852 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

countants would be out of the picture, as the plaintiff would

have been able to recover with interest all the sums that it ad-

vanced to the bankrupt party. But the question of Touche’s

negligence liability arose precisely because Stern was insolvent,

and thus the issue became whether the defendant accountant

should be liable for its inability to detect a third-party fraud by

exercising reasonable care in doing its audit. It was evident

that the accountants and the plaintiffs were not in privity with

each other, although it was perfectly evident that the audit

statement was intended to be used, and would be used, to ob-

tain loans from third parties like this plaintiff. Judge Cardozo

nonetheless denied the action by drawing a distinction be-

tween fraud and negligence:

To creditors and investors to whom the employer ex-

hibited the certificate, the defendants owed a like

duty to make it without fraud, since there was notice

in the circumstances of its making that the employer

did not intend to keep it to himself. A different ques-

tion develops when we ask whether they owed a duty

to these to make it without negligence. If liability for

negligence exists, a thoughtless slip or blunder, the

failure to detect a theft or forgery beneath the cover

of deceptive entries, may expose accountants to a lia-

bility in an indeterminate amount for an indetermi-

nate time to an indeterminate class.

148

This passage is notable for several key points. The first is

that Cardozo thinks that the ambit for liability in fraud is al-

ways narrower than negligence. In both settings, he is pre-

pared to go beyond the strict privity limitation found in Mac-

Pherson. He is willing to extend liability for fraud to those per-

sons to whom the client exhibited the audit report, for it was

well known that the document would be key in making a loan

or a sale, and would typically be transmitted without alteration,

although not necessarily without written or oral embellish-

ment. The power of that privity limitation remains largely in-

tact today, as efforts to expand the potential right of recovery

to the public at large,

149

or even to those whom the defendant

148. Id. at 179 (citations omitted).

149. See, e.g., White v. Guarente, 372 N.E.2d 315, 318 (N.Y. 1977) (confin-

ing liability to “a known group possessed of vested rights,” and not “the ex-

tensive and indeterminable investing public-at-large”).

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 853

might “reasonably foresee” might rely on the statement, have

largely faltered, given that it would impose enormous liability

upon parties, like accountants, who may receive but a modest

financial reward for their services.

150

The reason why courts are willing to go beyond privity in

fraud cases is that the defendant has it wholly within its power

to avoid making these wrongful statements. In cases like Ul-

tramares, however, where only negligence is at issue, the poten-

tial liability could unintentionally ratchet up, rapidly ex-

tending far and wide. Cardozo himself was well aware of the

difference. In Ultramares, citing MacPherson, he mused that

[A] manufacturer who is negligent in the manufac-

ture of a chattel in circumstances pointing to an un-

reasonable risk of serious bodily harm to those using

it thereafter may be liable for negligence though

privity is lacking between manufacturer and user. A

force or instrument of harm having been launched

with potentialities of danger manifest to the eye of

prudence, the one who launches it is under a duty to

keep it within bounds . . . . Even so, the question is

still open whether the potentialities of danger that

will charge with liability are confined to harm to the

person, or include injury to property. . . . In either

view, however, what is released or set in motion is a

physical force. We are now asked to say that a like

liability attaches to the circulation of a thought or a

release of the explosive power resident in words.

151

His answer to the question was no, and the explanation

rests on the different characteristics of goods and information.

As noted earlier, words can move with a rapidity that is not

possible with respect to physical forces, which in most cases

quickly come to rest. In ordinary physical-injury cases, the

number of joint-causation situations tends to be very small,

and so (for the most part) the extension of MacPherson from

personal injuries to property damage seems relatively safe.

Certainly, if the crash damaged the contents of the car, the

150. For details of that view, see Rosenblum v. Adler, 461 A.2d 138, 147 (N.J.

1983), superseded by statute,

N.J. S

TAT

. A

. § 2A:53A-25 (

West

2017),

which

was in turn rejected in Bily v. Arthur Young & Co., 834 P.2d 745, 767 (Cal.

1992).

151. Ultramares, 174 N.E. at 181 (citations omitted).

854 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

principle of liability would be about the same, and the exten-

sion to different types of personal property is little different

from the extension of tort liability to multiple persons in the

same car. It is also worth noting that, historically, MacPherson,

like Escola afterwards, had no impact whatever on business ac-

tivities given the remaining constraints on liability. It was only

when tort law was applied to toxic releases that it encountered

the hazards latent in seeking to impose damage remedies on

countless individuals, each of whom has contributed at most a

tiny fraction of loss to any one of countless other individu-

als.

152

The flow of information, however, differs radically from

the movement of goods, which helps explain the powerful eco-

nomic rationale that lies behind the restrictive rules in Ul-

tramares. In any face-to-face negotiations, the defendant must

receive compensation sufficient to cover the risk of tort liabil-

ity. Where negligence is the supposed norm, a principle like

res ipsa loquitur could easily allow a jury to infer negligence

from the fact of error, so that the system verges on strict liabil-

ity,

153

which may be fine in cases involving physical injury to

strangers, but is more dubious in medical malpractice cases,

where the strict liability principle has long been regarded as

too expansive for a sustainable market in medical or health-

related services, given the inherent riskiness of medical proce-

dures.

154

The market for medical services will shut down if the

liability in question is disproportionate to the payment re-

ceived to enter into the transaction in the first place, which is

why Cardozo, for example, was acutely aware of the need to

keep the total financial exposure proportionate to the consid-

eration received for the services rendered.

155

As he observed,

“A promisor will not be deemed to have had in mind the as-

152. See

ESTATEMENT

HIRD

)

ORTS

: A

PPORTIONMENT OF

IABILITY

§ 26 (A

. L

NST

. 2000)

;

ESTATEMENT

ECOND

)

ORTS

§ 433A (A

NST

. 1965) (

dealing with the apportionment problem).

153. For its various formulations, see

ESTATEMENT

ECOND

)

ORTS

§ 328D (A

. L

NST

. 1965); R

ESTATEMENT

HIRD

)

ORTS

: L

IABILITY FOR

HYSICAL AND

MOTIONAL

ARM

§ 17 (A

. L

NST

. 2010).

154. For an early statement of this rationale, see Clarence Morris, Custom

and Negligence, 42

OLUM

. L. R

1147, 1164–65 (1942).

155. See H.R. Moch Co. v. Rensselaer Water Co., 159 N.E. 896, 898–99

(N.Y. 1928) (dealing with the question of liability of a utility for the failure to

supply water pressure), cited in Ultramares, 174 N.E. at 448.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 855

sumption of a risk so overwhelming for any trivial reward.”

156

Nor, it should be added, should the promisee labor under any

such illusion that it can impose liability to remove or cushion

its loss. The common thread that comes through all these

cases is that a market system will break down if the liability

imposed by courts for adverse events is disproportionate to any

fee that the defendant had earned. So, the judgment in Ul-

tramares (and similar cases) is best understood as insisting that

the plaintiff could have hired its own auditor on terms that

would have, for a healthy premium, allowed it to recover for

any losses made on the advances, on whatever terms and con-

ditions the two parties agree.

157

The principle of freedom of

contract thus sneaks in through the back door to explain the

use of the privity doctrine in these cases of financial loss.

The narrow approach to privity and remoteness of dam-

age in Ultramares carried over to the instructive case of Holmes

v. Securities Investor Protection Corp.,

158

which imposed a direct

injury restriction for cases of financial loss, even when attribu-

table to fraud. Holmes was a stock-manipulation case in which it

was alleged that, between 1964 and 1981, the defendants made

unduly optimistic statements about six companies in which

they held stakes in order to drive their stock prices higher;

they also engaged in a number of sham trades in order to

make these shares appear more liquid than they were. On the

strength of these activities, the broker-dealers purchased suffi-

156. Moch, 159 N.E. at 898. For an endorsement of this view, see Charles

O. Gregory, Gratuitous Undertakings and the Duty of Care, 1

AUL

L. R

30,

60 (1951) (“Cardozo thought the sum of $42.50 insufficient to warrant the

conclusion that a negligent water company should be made to relieve a fire

insurance company from bearing the ultimate risk of loss by fire . . . .”). The

full explanation is a bit more complex, for it requires an explanation for why

the first-party market works better. It does so because it is easier to tie cover-

age to the harm suffered by the plaintiff than it is to those harms brought

about by a specific source, which may be difficult to identify in many cases.

The contract law also provides greater flexibility on such key issues as de-

ductibles and policy limits that cannot be introduced under a simple tort

regime.

157. Victor P. Goldberg, Accountable Accountants: Is Third-Party Liability Nec-

essary?, 17

J. L

EGAL

TUD

295, 301 (1988)(urging that the best solution is for

courts to “allow the parties to resolve the problem by contract.”). The point

has also been made in the early cases in this area. See, e.g., Cabot v. Christie,

42 Vt. 121 (1869).

158. Holmes v. Sec. Inv’r Prot. Corp., 503 U.S. 258 (1992).

856 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

cient quantities of the shares to lead to their financial ruin.

159

The Securities Investor Protection Corporation (SIPC) was

charged with protecting customers from financial loss when

the broker-dealers who held their accounts went bankrupt.

The SIPC made the appropriate payments to various custom-

ers, subrogated the customers’ rights against the defendants,

and then brought an action against Holmes and his co-con-

spirators. The SIPC alleged a pattern and practice of violations

of the Securities Act, which gave rise to liability under the

Racketeer Influenced and Corrupt Organizations Act

(RICO)

160

for participation in a stock-manipulation scheme

that was said to have “disabled” the broker-dealers from meet-

ing their obligations to their customers.

161

The basic substantive provision of RICO states:

Any person injured in his business or property by rea-

son of a violation of section 1962 of this chapter may

sue therefor in any appropriate United States district

court and shall recover threefold the damages he sus-

tains and the cost of the suit, including a reasonable

attorney’s fee.

162

In line with established case law, the Court held that “but

for” causation (always a dubious account) was not sufficient,

and that proximate cause had to be shown.

163

In this context,

the reliance on that dual combination meant that the Court

looked closely at the various decisions by other actors that in-

tervened between what the defendant did and what losses were

suffered by the broker-dealers, incorporating “common-law

principles of proximate causation.”

164

It then noted that even

if subrogation was accepted, tort liability did not follow be-

159. Id. at 262–63.

160. 18 U.S.C.A. §§ 1962, 1961(1) & (5) (West 1988).

161. Holmes, 503 U.S. at 261.

162. 18 U.S.C.A. § 1964(c) (West 1995).

163. Holmes, 503 U.S. at 268.

164. Id., with the Court’s near-obligatory citation to

W. K

EETON

, D. D

OBBS

R. K

EETON

& D. O

WEN

ROSSER AND

EETON ON

AW OF

ORTS

§ 41, at 264

(5th ed. 1984) (“Here we use ‘proximate cause’ to label generically the judi-

cial tools used to limit a person’s responsibility for the consequences of that

person’s own acts. At bottom, the notion of proximate cause reflects ‘ideas

of what justice demands, or of what is administratively possible and conve-

nient.’”). In my view, the principles of proximate causation are a good deal

more precise than this counsel of despair suggests. See

ICHARD

A. E

PSTEIN

ORTS

258–73 (1999).

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 857

cause the many factors apart from trading in these shares

could have led to the insolvency of the two broker-dealers. The

Court stated:

[T]he link is too remote between the stock manipula-

tion alleged and the customers’ harm, being purely

contingent on the harm suffered by the broker-deal-

ers. That is, the conspirators have allegedly injured

these customers only insofar as the stock manipula-

tion first injured the broker-dealers and left them

without the wherewithal to pay customers’ claims. Al-

though the customers’ claims are senior (in recourse

to “customer property”) to those of the broker-deal-

ers’ general creditors, the causes of their respective

injuries are the same: The broker-dealers simply can-

not pay their bills, and only that intervening insol-

vency connects the conspirators’ acts to the losses suf-

fered by the nonpurchasing customers and general

creditors. As we said, however, in Associated General

Contractors [of California, Inc. v. Carpenters],

165

quoting

Justice Holmes, “The general tendency of the law, in

regard to damages at least, is not to go beyond the

first step.”

166

Justice Holmes’s opinion just cited, in Southern Pacific Co.

v. Darnell-Taenzer Lumber Co.,

167

only adds to the confusion.

The difficulty is that Justice Holmes’s proposition about ordi-

nary tort law in physical injury cases was manifestly wrong. At

the time he wrote it, there was still some support for the so-

called “last wrongdoer” rule in physical injury cases,

168

but it

had by and large given way to a rule that said that if the first

defendant created a dangerous condition, which then resulted

in harm because of the subsequent act of a second defendant,

both could be held responsible for the harm, even if they had

not coordinated the activities among them.

169

The number of

165. Associated Gen. Contractors of Cal., Inc. v. Carpenters, 459 U.S. 519,

534 (1983) (quoting S. Pac. Co. v. Darnell–Taenzer Lumber Co., 245 U.S.

531, 533 (1918)).

166. Holmes, 503 U.S. at 271 (1992).

167. S. Pac. Co. v. Darnell–Taenzer Lumber Co., 245 U.S. 531, 533 (1918).

168. For its leading defense, see

HOMAS

EVEN

, N

EGLIGENCE IN

(3d ed. 1908).

169. See, e.g., Atherton v. Devine, 602 P.2d 634 (Okla. 1979). For the gen-

eral statements about intervening acts, see

ESTATEMENT

ECOND

)

ORTS

858 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

cases is small, and the dangerous condition in question—say

an uncovered hole in the street—had to persist until the time

of the injury. But Darnell-Taenzer was a rate-regulation case in

which the plaintiff charged that he was hurt because he had to

pay excessive amounts for a shipment, which in turn he could

recover from his customers. Holmes thought that this question

prompted the following answer:

The answer is not difficult. The general tendency of

the law, in regard to damages at least, is not to go

beyond the first step. As it does not attribute remote

consequences to a defendant so it holds him liable if

proximately the plaintiff has suffered a loss. The

plaintiffs suffered losses to the amount of the verdict

when they paid. Their claim accrued at once in the

theory of the law and it does not inquire into later

events.

170

The rule here is the soul of good sense, but only for ad-

ministrative reasons that are orthogonal to the theory of proxi-

mate causation. The immediate plaintiff is in a good position

to challenge the overpayment. But if there is a setoff for recov-

ery that the ratepayer receives from other parties, then one of

two things must happen. The first is that the next downstream

party, and the one after that, each have determinate losses

equal to a fraction of the overcharge. The second is that each

party up and down the chain must have an action for its actual

losses. In either case, the coordination problems among multi-

ple claimants are acute; one way or the other, it becomes nec-

essary to formulate estimates of the ultimate harm borne by

each injured party, which would nearly always consume more

resources, and involve more parties, than the whole case is

worth. Channeling all the losses to the immediate purchaser

eliminates these confusions while improving the incentives for

the plaintiff to sue to recover the amount of the overcharge.

171

Historically, the rule announced is not limited only to rate

§§ 431, 448 & 449 (

. L

NST

. 1965). The Third Restatement does not

trouble itself with any of these refinements when it adopts broad tests in

cases of intervening actions and supervening causes.

ESTATEMENT

HIRD

)

ORTS

§ 34 (

. L

NST

. 2013). For a comprehensive account, see

P-

STEIN

supra note 164.

170. Darnell-Taenzer, 245 U.S. at 533–34.

171. On the importance of channeling, e.g., limiting the number of plain-

tiffs in torts cases, see William Bishop, Economic Loss in Tort, 2

XFORD

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 859

cases. Indeed, it was explicitly carried over to treble-damage

antitrust cases in Illinois Brick Co. v. Illinois, which treated

Darnell-Taenzer as a “particularly apt precedent.”

172

These ad-

ministrative imperatives displace, rather than incorporate, a

common-law proximate-cause analysis, although leading to a

similar conclusion.

It is therefore easy to distinguish Holmes from its key pre-

cedent. Nonetheless, the decision makes perfectly good sense,

because of the problem of deciding how the purchase of these

inflated stocks mixed in with other management mistakes that

the broker-dealers may well have made. The ultimate lesson

thus remains the same. The complex patterns of interaction of

various decisions made on information that comes from multi-

ple sources tend to render the tort action difficult, but not

impossible, to sustain. It is therefore an open question

whether the action should apply if SIPC could prove that huge

purchases made in explicit reliance on the fraud, with all

other forms of mismanagement ruled out, caused the losses.

But excluding other causes, which is a staple of res ipsa loquitur

in physical injury cases, is much harder to accomplish as the

class of potential causes rapidly increases.

173

Holmes has therefore been followed repeatedly whenever

it is hard to disentangle multiple influences on a given out-

come. To give but one other RICO example from the Second

Circuit, Laborers Local 17 Health and Benefit Fund v. Philip Morris,

Inc. asked whether the plaintiff health and benefit fund could

recover from Philip Morris for the additional costs that it had

to bear in treating patients who had contracted additional ill-

nesses from smoking.

174

Plaintiff alleged that Philip Morris en-

gaged in fraudulent conduct that induced its union members

to overconsume cigarettes. The RICO case followed closely on

EGAL

TUD

. 1 (1982); Mario J. Rizzo, A Theory of Economic Loss in the Law of

Torts, 11

J. L

EGAL

TUD

. 281, 305 (1982).

172. Ill. Brick Co. v. Illinois, 431 U.S. 720, 751 (1977).

173. See, e.g.,

ESTATEMENT

ECOND

)

ORTS

§ 328D (

. L

NST

1965), which refers to “(b) other responsible causes, including the conduct

of the plaintiff and third persons, are sufficiently eliminated by the evi-

dence.” The test in Prosser & Keeton on Torts requires that the accident “(2)

must be caused by an agency or instrumentality within the exclusive control

of the defendant.”

OBBS

, K

EETON

& O

WEN

, supra note 164, at 244.

174. Laborers Local 17 Health & Benefit Fund v. Philip Morris, Inc., 191

F.3d 229 (2d Cir. 1999).

860 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

the heels of Holmes and likewise foundered on the proximate-

cause questions that made it impossible to disentangle various

causes:

[P]laintiffs’ alleged damages might have derived

from inefficiencies in the Funds’ own management,

as well as from non-smoking related health problems

suffered by the smokers, and it would be the sheerest

sort of speculation to determine how these damages

might have been lessened had the Funds adopted the

measures defendants allegedly induced them not to

adopt. The complexity of these calculations makes

the ultimate question of damages suffered by the

Funds virtually impossible to determine.

175

To this list can be added the question of how much each

individual plaintiff knew about the risk at various times in his

or her smoking career, and what other physical exposures

could have contributed to their underlying condition. The

causal chain is indeed filled with missing links, which explains

why the relatively generous rules on causation followed in

physical-injury cases are rejected in connection with financial

losses. The analysis here is perfectly general and applies with

equal force to common-law cases, securities-act cases, RICO

cases, and, of course Martin Act cases. The same cautious atti-

tude we saw in the privity context carries over to cases in which

the fraud is made in connection with a public offering of se-

curities, where the number of potential purchasers is far

greater.

The logic of Ultramares proved highly influential in the im-

portant Supreme Court decision in Blue Chip Stamps v. Manor

Drug Stores.

176

Blue Chip was not a common-law action, but re-

lied instead on Section 10(b) of the Securities Act of 1934,

177

as interpreted under Rule 10b-5, first promulgated in 1948.

178

175. Id. at 240.

176. Blue Chip Stamps v. Manor Drug Stores, 421 U.S. 723, 747–48 (1975)

(quoting Ultramares).

177. 15 U.S.C. § 78j(b) (2012).

178. 17 C.F.R. § 240.10b-5 (2017). The text of the regulation in full reads:

§ 240.10b-5 Employment of manipulative and deceptive devices.

It shall be unlawful for any person, directly or indirectly, by the

use of any means or instrumentality of interstate commerce, or of

the mails or of any facility of any national securities exchange,

(a) To employ any device, scheme, or artifice to defraud,

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 861

The case posed the simple question of whether a party who

was neither a buyer nor seller of shares could maintain an ac-

tion for damages by alleging that the defendant seller of the

shares had “made the prospectus overly pessimistic in order to

discourage respondent and other members of the allegedly

large class whom it represents from accepting what was in-

tended to be a bargain offer.”

179

As a matter of basic incen-

tives, the reliance associated with a decision not to purchase

introduces the same distortion into the market as the converse

situation in which an unwise purchase or sale is made on the

strength of false statements. Therefore, if administrative con-

cerns did not loom large in the overall equation, the rough

economic equivalence between acting and not acting on ei-

ther the buy or sell side should make both equally relevant.

There were, however, two instructive responses to this

damage action, one statutory and the other administrative.

First, liability for transactions that did not take place because

of the defendant’s misrepresentations is not covered by the

key words in the federal statute, which limits coverage “in con-

nection with the purchase or sale of any security.”

180

As Justice

Powell noted in his concurrence, the clause in question did not

read “in connection with the purchase or sale of, or an offer to

sell, any security.”

181

That limitation is not relevant with re-

spect to the Martin Act, which applies to these transactions “re-

gardless of whether issuance, distribution, exchange, sale, ne-

gotiation or purchase resulted.”

182

As Justice Rehnquist’s

noted, twice before, in both 1957 and 1959, the SEC had re-

quested that Congress amend Section 10(b) to read: “in con-

nection with the purchase or sale of, or any attempt to purchase or

sell, any security.”

183

(b) To make any untrue statement of a material fact or to omit

to state a material fact necessary in order to make the state-

ments made, in the light of the circumstances under which

they were made, not misleading, or

operates or would operate as a fraud or deceit upon any per-

son, “in connection with the purchase or sale of any security.

179. Blue Chip Stamps, 421 U.S. at 726–27.

180. 15 U. S. C. § 78j(b) (2012). 17 C.F.R. § 240.10b-5 (2017).

181. Blue Chip Stamps, 421 U.S. at 756 (Powell, J., concurring).

182. N.Y. G

. B

§ 352-c(1)(c).

183. Blue Chip Stamps, 421 U.S. at 732.

862 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

Yet, when those words were put into the 1921 Martin Act,

the sole remedy was injunctive relief, and it made perfect

sense to allow the stoppage of these offerings before they

could work harm on ordinary individuals if left uncorrected.

Here, the governing principle in the fraud context is no differ-

ent from that governing the various kinds of property damage

attributable to common-law nuisances: It often makes sense to

allow the state to enjoin a nuisance before it occurs, rather

than to have to wait for the occurrence of the harm. But the

same caveat applies in both cases as well. The definitions of

fraud and nuisance should be appropriately cabined to ensure

that prophylactic relief does not lead to some massive abuse

that suppresses freedom of expression, in the one case, and

freedom of action, including the use and development of real

property, in the other.

Given these basic considerations, it should come as no

surprise that the SEC also has statutory power to issue injunc-

tive relief under Section 20(b) of the 1933 Securities Act for

violations of Section 17(a)

184

of that same law, which applies

only to sellers who engage in the use of fraud or other manip-

ulative devices by sellers “in the offer or sale of any securi-

ties,”

185

authorizing both damages and injunctive relief.

186

184. Section 17(a) of the 1933 Act provides:

It shall be unlawful for any person in the offer or sale of any securi-

ties . . . by the use of any means or instruments of transportation or

communication in interstate commerce or by use of the mails, di-

rectly or indirectly—

(1) to employ any device, scheme, or artifice to defraud, or

(2) to obtain money or property by means of any untrue state-

ment of a material fact or any omission to state a material fact

necessary in order to make the statements made, in light of the

circumstances under which they were made, not misleading,

(3) to engage in any transaction, practice, or course of busi-

ness which operates or would operate as a fraud or deceit

upon the purchaser.

Securities Act of 1933 § 17(a), 48 Stat. 84 (codified as amended at 15 U.S.C.

§ 77q(a) (2010)).

185. Id.

186. Section 20(b) of the 1933 Act provides:

Whenever it shall appear to the Commission that any person is en-

gaged or about to engage in any acts or practices which constitute

or will constitute a violation of the provisions of this subchapter [e.

g., § 17 (a)], or of any rule or regulation prescribed under author-

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 863

Similarly, the SEC has the power under Section 21(d)

187

of the

1934 Act to seek injunctive relief against a person who is “en-

gaged or is about to engage in acts or practices constituting a

violation” of the Act, including Section 10(b),

188

and the regu-

lations promulgated thereto under Rule 10b-5.

189

In dealing

with this fourfold set of statutory provisions, Congress (at least

as construed by the Supreme Court) divided the baby by find-

ing that scienter was not required for violations of Section

17(a) of the 1933 Act, but was required for injunctive relief

under Section 10(b) of the 1934 Act.

190

In Aaron v. SEC, the

Supreme Court, relying largely on the legislative history, held

that scienter was required for both injunctive relief and for

damages under Section 10(b) and Rule 10b-5,

191

but it issued

a divided verdict under Section 17(a) of the 1933 Act, finding

that scienter was required in damage actions under Section

17(a)(1) but not under Section 17(a)(2).

192

The latter prohib-

its any person from obtaining money or property “by means of

any untrue statement of a material fact or any omission to state

a material fact”

193

and contains no scienter requirement at all.

The Court also reached the same conclusion with respect to

Section 17(a)(3),

194

which made it unlawful “to engage in any

transaction, practice, or course of business which operates or

would operate as a fraud or deceit,”

195

and was directed to-

ward effects and not intentions. It seems quite clear that the

differences between the Securities Acts and the Martin Act on

injunctive relief are on the smaller side, especially since it is

always easy to establish scienter for a defendant who continues

ity thereof, it may in its discretion, bring an action in any district

court of the United States . . . to enjoin such acts or practices, and

upon a proper showing a permanent or temporary injunction or

restraining order shall be granted without bond.

Securities Act of 1933 § 20(b), 48 Stat. 86 (codified as amended at 15 U.S.C.

§ 77t(b) (2010)).

187. Securities Exchange Act of 1934 § 21(d), 48 Stat. 900 (codified as

amended at 15 U.S.C. § 78u(d) (2015)).

188. 15 U.S.C. § 78j(b).

189. 17 C.F.R. § 240.10b-5 (1975).

190. Aaron v. SEC, 446 U.S. 680 (1980).

191. Id. at 691.

192. Id. at 697.

193. Id. at 687.

194. Id. at 697.

195. Id. at 687.

864 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

to engage in certain practices after being warned of their ille-

gality.

The central point of principle, however, is that it is gener-

ally unwise to authorize individual damage actions against po-

tential buyers and sellers, which was the point raised in Blue

Chip. On this point, the attitude taken in Ultramares invited the

Court in Blue Chip to stress the key points of principled differ-

ence between the action given to defrauded buyers and sellers

and that action which might be given to other persons who did

not go forward with the purchase or sale. Of these differences,

two are key. The first is the potential scope of the action. The

requirement of an actual transaction limits the number of po-

tential parties to the suit. An offering to the public that under-

states potential value could drive away countless numbers of

individuals, when it is never clear whether they would have

purchased or sold even if provided with full and adequate in-

formation. These potential plaintiffs have the benefit of hind-

sight so that the disappointed purchasers make their claims

only for those stocks that have soared after the original offer-

ing. Otherwise, they remain silent. Second, the absence of a

specific transaction and interaction makes it difficult to prove

the fact of reliance, rendering it difficult to determine the ap-

plicable measure of damages.

196

Any current shareholder that

neither buys nor sells is always a potential plaintiff once the

class of putative plaintiffs is expanded to cover at least some

parties who engage in no action.

It is therefore no wonder that the Supreme Court shrank

from giving the normative theory of deceit full range in cases

rich with institutional complications. In cases like these, it

must be remembered that direct administrative oversight by

the SEC can be imposed without the remedial complications

involved in private rights of action, which makes it sensible to

withhold private liability even when some form of government

action is warranted. And what makes sense in connection with

the common law and the federal securities law also makes

sense in connection with the Martin Act. Indeed, the New York

courts have worked an effective integration between regulatory

relief and private rights of action. In CPC International Inc. v.

McKesson Corp., a divided New York Court of Appeals had held

196. See Blue Chip Stamps v. Manor Drug Stores, 421 U.S. 723, 735–39

(1975) (discussing these points).

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 865

that no private right of action could be implied under the Mar-

tin Act.

197

The result paralleled that under Section 17(a) of

the Securities Act, which, the New York Court of Appeals again

held, implied no private right of action.

198

Once again the par-

allels between the state and federal law are more significant

than any differences on the same point. Yet by the same token,

in Assured Guaranty (UK) Ltd. v. J.P. Morgan Investment Manage-

ment Inc., the New York Court of Appeals held that the Martin

Act did not preempt private rights of action for breach of fidu-

ciary duty and for gross negligence in the management of

portfolio assets, both non-fraud claims, for which the plaintiff

relied on a guarantee.

199

Again the result is exactly correct,

because there is no reason why the regulatory oversight should

block a garden-variety private action brought under a contract

of guarantee.

200

Instructively, Assured Guaranty did not cite to

Ultramares, because the case itself did not challenge any of the

outer limits on common-law actions for fraud that were impli-

cated in that case.

The same set of concerns about rapid information flow

also appear in the law of insider trading when the information

that comes from an unauthorized release by the insider travels

through the hands of multiple parties and is combined with

information that comes from other sources.

201

The point was

of special urgency in the highly publicized decision in United

States v. Newman,

202

where the alleged tippees were three or

four steps removed from the insiders who first divulged the

information. As that information passed through the hands of

these intermediaries before being, at last, received by the de-

fendants, it was combined with other information that made it

197. CPC Int’l Inc. v. McKesson Corp., 514 N.E.2d 116, 118 (N.Y. 1987).

198. Id. at 122.

199. Assured Guar. (UK) Ltd. v. J.P. Morgan Inv. Mgmt. Inc., 962 N.E.2d

765, 771 (2011).

200. See id. at 769 (“Here, the plain text of the Martin Act, while granting

the Attorney General investigatory and enforcement powers and prescribing

various penalties, does not expressly mention or otherwise contemplate the

elimination of common-law claims . . . . Certainly the Martin Act, as it was

originally conceived in 1921 with its limited [injunctive] relief, did not

evince any intent to displace all common-law claims in the securities field.”).

201. For my extensive treatment of these issues, see Epstein, supra note 87.

Note that the use of the common-law reference point is as appropriate for

insider trading cases as for misrepresentation cases. Id.

202. United States v. Newman, 773 F.3d 438 (2d Cir. 2014).

866 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

unclear how much of the information came from improper

inside sources and how much came from legitimate ones. In

the absence of solid evidence on the flow of the information,

the action foundered, as it should have. “[N]o rational jury

would find that the tips were so overwhelmingly suspicious

that Newman and Chiasson [a codefendant] either knew or

consciously avoided knowing that the information came from

corporate insiders or that those insiders received any personal

benefit in exchange for the disclosure.”

203

Indeed, this issue is more complicated for another reason.

The securities law is at its greatest strength when the parties

who have given away or relied on information are acting in

breach of a voluntarily assumed fiduciary duty. For instance, in

United States v. O’Hagan,

204

a partner in a law firm used confi-

dential information about their client’s forthcoming tender of-

fer to purchase stock and cash out when the offer became pub-

lic, a clear breach of fiduciary duty to that client. But in the so

called classical theory of insider trading endorsed in

O’Hagan,

205

the basis of the duties that officers and directors

owe to shareholders is much more obscure: the standard rests

on positive law, not voluntary arrangement, and thus is far

more likely to be inefficient relative to what private parties

would desire in face-to-face negotiations. Indeed, one diffi-

culty in the plaintiff’s entire theory in Newman was that the

insider information was released in gray-market transactions,

where the managers of the firms wanted to get the informa-

tion out quietly, even selectively, in order to stimulate interest

in the stock. It was only the existence of a strong nondiscrimi-

nation rule imposed by the SEC in Fair Disclosure, Regulation

206

that made it possible to be completely above-board

about those releases. But those cases stand in sharp contrast to

203. Id. at 455. Dirks v. SEC, 463 U.S. 646, 662 (1983) (The last clause

refers to the misguided requirement that there be a return benefit to the

tippee for the receipt of the information to count, I regard that so-called

benefit requirement as wholly irrelevant). See also Epstein, supra note 87, at

1504–10 (the proper analogy is to the receipt of stolen goods, where the

good-faith donee does not prevail over the true owner. It is as wrong to trade

on donated information as purchased information).

204. U.S. v. O’Hagan, 521 U.S. 642 (1997).

205. See id. at 651–52.

206. Final Rule: Selective Disclosure and Insider Trading, SEC Release

No. 33-7881, 17 C.F.R. §§ 240, 243, 249 (proposed Dec. 20, 1999), https://

www.sec.gov/rules/final/33-7881.htm.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 867

others in which the information that is passed on to third par-

ties has been misappropriated from the company. In this last

setting, the information should be treated as stolen property,

which potentially makes everyone involved in the release, or in

receipt of the information as a tippee, both civilly and crimi-

nally liable.

207

As these materials show, the proper approach is sensitive

to transaction-costs considerations. Thus when harms are wide-

spread, enforcement of the basic rules is better left to direct

government action than to individual suits. The key point,

however, is that the shift from private to public enforcement

should only be used to reduce the transaction cost of enforce-

ment. It is not an open invitation for legislatures or courts to

expand the definition of harm so as to create a decisive break

between the public and private law, thereby generating the

endless liability that Cardozo sought to avoid, especially in

these financial negligence cases.

208

The sky is the limit on gov-

ernment action if administrative agencies can “redefine” prop-

erty rights at will, without having to pay just compensation for

the property interest taken. The constant cross-references be-

tween common-law rules, the federal securities acts, and the

Martin Act rests on the simple fact that they all address the

same fundamental problem of how best to deal with informa-

tion breakdown. Therefore, to the extent the Martin Act elimi-

nates the scienter problem, it contains the seeds of its own mis-

application whenever, as is sometimes the case, it expands the

potential scope of liability in ways that the traditional law of

misrepresentation, including privity limitation, rightly rejects.

III.

ALLEABLE

ARTIN

CT IN

CTION

The next task is to examine the notable cases that have

been brought under the Martin Act. As the previous discussion

indicates, this job cannot be undertaken in a vacuum, because

207. See Salman v. United States, 137 S. Ct. 420 (2016). For my commen-

tary, see Richard A. Epstein, Intellectual Myopia On Insider Trading,

OOVER

NST

.: D

EFINING

DEAS

(Oct. 10, 2016), http://www.hoover.org/research/in

tellectual-myopia-insider-trading.

208. See Richard A. Epstein, From Common Law to Environmental Protection:

How the Modern Environmental Movement Has Lost Its Way,

23 S

. C

. E

CON

. 141

, 143 (2015).

868 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

it is critical to note the similarities and differences between

Martin Act cases, on the one hand, and common-law actions

for deceit and federal securities-fraud actions, on the other. I

shall concentrate on three such situations here. The first is the

original Martin Act prosecution of bogus financial schemes.

The second is the Act’s use in the more complicated financial

transactions of AIG and its then-Chairman Maurice “Hank”

Greenberg and then-CFO Howard Smith. The third involves

the now fizzling campaign of former NYAG Schneiderman to

use the Martin Act against ExxonMobil on charges of climate

fraud. The overall verdict is that the early cases gave rise to

little social concern, at least on the basic matters of liability.

Indeed, the NYAG’s case against Greenberg and Smith was

sound, but only on the issue of liability, given their admissions

of scienter. But—by the same token—the decision to force his

resignation was on balance a classic example of government

overreach, just as it would have been overreach under the fed-

eral securities law. At the same time, the ongoing case that

Schneiderman brought against ExxonMobil for improper con-

duct was as unsound and ill-conceived under the Martin Act,

both on questions of liability and damage, as it would have

been under the federal securities law.

A. Busting Bogus Financial Schemes.

The initial use of the Martin Act was to attack routine fly-

by-night schemes of financial fraud. Judges perceived the basic

scope of the social problem, and in 1926 they consciously and

unanimously opted for a broad reading of the statute in New

York v. Federated Radio Corp.,

209

a case involving only injunctive

relief. In opposing that decision, the defendants argued that a

broad definition of fraud ran afoul of the Due Process Clause:

“[The Martin Act] does not define the ‘fraud’ or the ‘fraudu-

lent practices’ which it purports to restrain. No enlargement

of the meaning of the act by implication being permissible, it

must be held to be applicable only to cases involving so-called

legal fraud according to the hitherto accepted definition.”

210

There is, in fact, a long tradition arguing for the narrow con-

struction of statutes passed in derogation of the common

209. People v. Federated Radio Corp., 244 N.Y. 33 (1926).

210. Id. at 33.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 869

law.

211

Nonetheless, Judge Cuthbert Pound brushed this aside

by reverting to the Supreme Court decision in Hall v. Geiger-

Jones Corp.,

212

which espoused a far broader proposition:

The purpose of the law is to prevent all kinds of fraud

in connection with the sale of securities and com-

modities and to defeat all unsubstantial and visionary

schemes in relation thereto whereby the public is

fraudulently exploited. The words ‘fraud’ and ‘fraud-

ulent practice’ in this connection should, therefore,

be given a wide meaning so as to include all acts, al-

though not originating in any actual evil design or

contrivance to perpetrate fraud or injury upon

others, which do by their tendency to deceive or mis-

lead the purchasing public come within the purpose

of the law.

213

From our evaluative perspective, the key question is

whether this broader definition creates risks that should have

been guarded against in the context of this case. The answer is

no. The object of the litigation in Federated was that “the adver-

tising matter of the defendants, contained in a promoters’ pro-

spectus, unduly inflates the value of the stock and conceals cer-

tain facts in regard to such values which should in good faith

be revealed.”

214

Given the private preparation of these selling

materials, it is highly doubtful that any one of the abundant

misstatements of fact contained therein were not deliberately

incorporated by the defendants as part of their sham selling

scheme. It is also highly doubtful that they could have credibly

contested their unique knowledge of the details of the venture

when they had sole control over the preparation of the sales

spiels from start to finish.

Moreover, the decision in Federated Radio Corp. puts no

pressure on the distinction between matters of fact and mat-

ters of opinion. The willingness to hold defendants responsi-

ble for culpable omissions is also consistent with basic tort the-

211. See, e.g., Roscoe Pound, The Common Law and Legislation, 21

ARV

. L.

383, 386 (1908) (“We are told commonly that three classes of statutes

are to be construed strictly: penal statutes; statutes in derogation of common

right; and statutes in derogation of the common law.”).

212. Hall v. Geiger-Jones Co., 242 U.S. 539 (1917).

213. Federated Radio Corp., 244 N.Y. at 38–39.

214. Id. at 40 (internal citations omitted).

870 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

ory. Finally, the likelihood that some targets would rely on the

prospectus, if circulated, should be taken as a given, as poten-

tial purchasers were the prospectus’s sole intended audience.

Aiming for an injunction prior to purchase prevents the need

to sort out thorny questions of causation once the false infor-

mation is unleashed to the defendant’s target audience. The

statement of Judge Pound also contains the words “in connec-

tion with the sale of securities”

215

and thus by implication ex-

presses the need to limit the potential ambit of the case, even

if under the Martin Act the context behind injunctive relief is

arguably broader than it was construed in some part of Blue

Chip Stamps.

216

Since the injunction prevents the sale, it

removes the need to think about the refund, the purchase

price or the loss of reliance damages when the deal goes

south. Nor did the broad reading of the statute invite system-

atic abuse in enforcement, as long as it is confined to en-

joining the phony stock schemes to which it was originally ap-

plied. The broad statute works well in this context, even if, as

will become evident, it does not work equally well in all others.

The same kind of analysis applies with equal force to most of

the condominium conversion cases noted above.

217

B. The AIG–Greenberg Litigation

As noted earlier in Section II.A,

218

there is nothing in the

text of the Martin Act that limits its application to the blue-sky

fraudulent offerings cases that spurred its adoption. Prior to

Eric Schneiderman, the most aggressive use of the Martin Act

was by then-NYAG Elliot Spitzer, who, it was said at the time,

rejected the “unspoken gentleman’s agreement” that ex-

empted “the big boys” from the Martin Act, by limiting “ac-

ceptable targets” to the likes of “shady pharmacists, Ponzi

schemes, and peddlers of fraudulent Salvador Dali litho-

graphs.”

219

Among the biggest targets that Spitzer attacked

were AIG, its former CEO Maurice “Hank” Greenberg, and

Howard Smith, its CFO during the critical years between 2001

215. Id. at 38.

216. See discussions supra Sections II.D.4, II.E.

217. See supra notes 73–77 and accompanying text.

218. See supra notes 53–54 and accompanying text.

219. Nicholas Thompson, The Sword of Spitzer,

EGAL

(May–June

2004), https://www.legalaffairs.org/issues/May-June-2004/feature_thomp

son_mayjun04.msp.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 871

and 2004. Thus, the original litigation covered a multitude of

transactions, prosecuted under both common-law rules of de-

ceit and the Martin Act.

Two transactions were key. The first of these, which took

place between 2000 and 2004, was between AIG and General

Re, a reinsurance carrier, while AIG was under the control of

Greenberg and his codefendant, Smith.

220

The gist of New

York’s claim was that the transaction in question was a sham

because it did not transfer any real risk to General Re, but was

designed solely to improve AIG’s balance sheet in ways that

allowed it to increase the stated reserves on its financial state-

ments.

221

That change was in turn alleged to have allowed AIG

to present a stronger position to public markets and thus enter

into more financial transactions.

222

The second of the dis-

puted transactions was with CAPCO, a company financed by

AIG, in which AIG is alleged to have used CAPCO to convert

underwriting losses into investment losses and thus present a

rosier financial picture to analysts. In the opening act of this

drama, Spitzer had forced out Greenberg as Chairman and

CEO of AIG and Smith as its CFO in 2005, about three years

before the company experienced serious financial losses in the

meltdown of 2008 under new financial management. Much of

the litigation brought against Greenberg and Smith disap-

peared in the years after that initial encounter.

223

By 2006,

with Greenberg and Smith out, new management quickly set-

tled those claims for $1.6 billion. One can criticize the size of

the settlement, but there is no dispute that AIG was, under

standard principles of vicarious liability, responsible for the ac-

tions of its two top officials.

The litigation against Greenberg and Smith went on for

another dozen years, as the defendants sought to derail the

prosecution. During the litigation, the NYAG dismissed all

common law fraud charges and dropped from his Martin Act

account all the transactions except for two that will be de-

scribed shortly. In People v. Greenberg,

224

Greenberg and Smith

220. See People v. Greenberg, 994 N.E.2d 838 (N.Y. 2013).

221. See id. at 840.

222. See id.

223. See Bethany McLean, Hank Greenberg’s Other AIG Trial: A Fight That Just

Won’t End,

ORTUNE

(Aug. 27, 2015), http://fortune.com/2015/08/27/

hank-greenbergs-other-aig-trial-a-fight-that-just-wont-end/.

224. People v. Greenberg, 946 N.Y.S.2d 1 (App. Div. 2012).

872 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

first sought to dismiss the charges on preemption grounds, in-

voking Title I of the Securities Litigation Uniform Standards

Act of 1998 (SLUSA),

225

the Private Securities Litigation Re-

form Act of 1995,

226

and the National Securities Markets Im-

provement Act of 1996.

227

But whether the matter turned on

express or implied preemption, there was no evidence whatso-

ever that these statutes, designed to recalibrate the operation

of the federal securities laws, had an unspoken collateral in-

tention to upend the long-established system of dual enforce-

ment that had been in place from the initial adoption of state

blue-sky laws, which predated the federal securities laws by fif-

teen years.

228

The court rebuffed Greenberg and Smith’s sum-

mary judgment request on the underlying claim. In so doing,

it did not rely on any of the strict liability strands of the Martin

Act, but instead invoked the standard rule under which “of-

ficers and directors are liable for a corporation’s fraud where

they either personally participate in the fraud or have actual

notice of its existence.”

229

As Greenberg and Smith had jointly

conducted both sets of negotiations, this participation stan-

dard was easily met.

The litigation continued for a further five years, but on

February 10, 2017, the case finally settled short of trial in part

due to the intercession of well-known mediator Kenneth Fein-

berg.

230

At the time of the settlement, Schneiderman issued a

public statement announcing, “Greenberg and Smith agree to

225. Securities Litigation Uniform Standards Act of 1998, Pub. L. No. 105-

353, § 101, 112 Stat. 3230 (codified as amended in scattered sections of 15

U.S.C. (2006)).

226. Private Securities Litigation Reform Act of 1995, Pub. L. No. 104-67,

109 Stat. 737 (codified as amended in scattered sections of 15 U.S.C.

(2006)).

227. National Securities Markets Improvement Act of 1996, Pub. L. No.

104-290, 110 Stat. 3416 (codified as amended in scattered sections of 15

U.S.C. (2006)).

228. People v. Greenberg, 994 N.E.2d 838, 840 (N.Y. 2013). Greenberg

and Smith sought certiorari in the preemption issue in Greenberg v. New York,

No. 16-294, filed August 30, 2016. It was denied on December 12, 2016. Peo-

ple v. Greenberg, 54 N.E.3d 74 (N.Y. 2016), cert. denied sub nom. Greenberg v.

New York, 137 S. Ct. 591 (2016). For relevant documents, see

SCOTUS

BLOG

http://www.scotusblog.com/case-files/cases/greenberg-v-new-york/ (last

visited Aug. 16, 2018).

229. Greenberg, 946 N.Y.S.2d at 9.

230. Randall Smith, Former A.I.G. Executives Reach Settlement in Accounting

Fraud Case,

N.Y. T

IMES

(Feb. 10, 2017), http://www.nytimes.com/2017/02/

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 873

return multi-million dollar bonuses they received while the

frauds were on AIG’s books.”

231

The statement also noted that

this settlement followed on the heels of an earlier settlement

that Greenberg and Smith had made with the SEC resulting in

the disgorgement of virtually all the payments made during

the 2001 to 2004 period when the transactions remained on

the books uncorrected. The wording of Schneiderman’s an-

nouncement tip-toed around the question of whether Green-

berg and Smith had personally committed fraud, even though

they had orchestrated the challenged transactions. The follow-

ing passage sums up the state of play:

If, as the parties presently concede, there was no risk

of loss in the [General Re] transaction, it should have

been recorded on AIG’s financials as a deposit. In-

stead, AIG recorded $250 million in loss reserves for

the fourth quarter of 2000 based upon the [General

Re] transaction and an additional $250 million in

loss reserves for the first quarter of 2001, consistent

with Greenberg’s intent when he reached out to Fer-

guson, to shore up the reserves. Had these amounts

not been credited in this manner, AIG would have

had a $187 million decline in its loss reserves by the

first-quarter of 2001.

232

The history of the CAPCO transaction was similarly sus-

pect: “After Greenberg and Smith left the company in 2005,

AIG announced that CAPCO involved an improper account-

ing structure created to characterize underwriting losses relat-

ing to the auto warranty business as capital losses.”

233

The

terms of the settlement called for Greenberg to pay $9 million

plus interest and for Smith to pay $900,000 with interest—part

of the money that the two men had received as performance

bonuses for the years between 2001 and 2004, when the al-

leged frauds had taken place. This settlement accompanied

Greenberg and Smith’s deals with the SEC, with the combined

10/business/dealbook/former-aig-executives-reach-settlement-in-account

ing-fraud-case.html?mcubz=0.

231. Press Release, N.Y. St. Att’y Gen., A.G. Schneiderman Announces Set-

tlement of Martin Act Case Against Former AIG CEO Maurice R. Greenberg

and Former AIG CFO Howard I. Smith (Feb. 10, 2017).

232. Greenberg, 946 N.Y.S.2d at 3–4.

233. Id. at 5.

874 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

disgorgement equaling the bonuses the defendants collected

during the 2001–2004 period. Schneiderman then pro-

claimed:

Today’s agreement settles the indisputable fact that

Mr. Greenberg has denied for twelve years: that Mr.

Greenberg orchestrated two transactions that funda-

mentally misrepresented AIG’s finances. After over a

decade of delays, deflections, and denials by Mr.

Greenberg, we are pleased that Mr. Greenberg has

finally admitted to his role in these fraudulent trans-

actions and will personally pay $9 million to the State

of New York.

234

The simultaneous statement by Greenberg did not con-

tain the word “fraud” but did contain the following statements:

The General Re transaction was done for the purpose

of increasing AIG’s loss reserves, and the Capco

transaction was done for the purpose of converting

underwriting losses into investment losses. I knew

these facts at the time that I initiated, participated in

and approved these two transactions.

For the years 2000 through 2002 (in the case of the

Capco transaction) and 2000 through 2003 (in the

case of the [General Re] transaction), I certified

AIG’s publicly-filed annual consolidated financial

statements aware that the financial effects of these

transactions were and continued to be reflected in

those statements.

As a result of these transactions, AIG’s publicly-filed

consolidated financial statements inaccurately por-

trayed the accounting, and thus the financial condi-

tion and performance for AIG’s loss reserves and un-

derwriting income. The accounting for the [General

Re] transaction was correctly restated by AIG in AIG’s

2005 Restatement of the Company’s financial results

for the years 2000 through 2003 and the first three

quarters of 2004. The accounting for the Capco

transaction was also restated by AIG in AIG’s 2005 Re-

statement of the Company’s financial results.

235

234. See Press Release, N.Y. St. Att’y Gen., supra note 231.

235. Statement, Maurice R. Greenberg (Feb. 10, 2017), https://ag.ny.gov

/sites/default/files/2017_02_10_greenberg_statement_final_2.pdf. See also

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 875

The SEC and Martin Act suits against Greenberg and

Smith were not the only lawsuits that arose out of these trans-

actions. In addition, the SEC ran its own investigation of the

transaction that yielded positive results: “In June 2005, two

[General Re] executives pleaded guilty to participating in a

conspiracy to commit securities fraud for their role in the

[General Re] transaction. In February 2008, four other [Gen-

eral Re] executives were convicted on federal criminal charges

with respect to the [General Re] transaction.”

236

The latter

convictions were reversed for a new trial on evidentiary

grounds, as one of the key witnesses, Richard Napier, a senior

General Re executive, “may well have testified falsely” on ma-

jor issues in the earlier federal case,

237

which effectively side-

lined him from any role in the prosecution under the Martin

Act. The Wall Street Journal made much of the inconsistencies

in his testimony, reflecting poorly on the NYAG.

238

But the

point is a sideshow. One can look into the disputed transac-

tion’s structure on its face by assessing whether AIG trans-

ferred sufficient risk to turn it into a legal transaction. Given

the admissions from Greenberg and Smith, it is apparent that

no risk was transferred. They rightly entered into the settle-

ment, in which they admitted scienter, because they knew that

the case against them was strong regardless of the disputed

and inconsistent Napier testimony.

On this score, it is worth noting that the original com-

plaint by Spitzer included other counts, both at common law

Statement, Howard I. Smith (Feb.10, 2017), https://ag.ny.gov/sites/default

/files/2017_02_10_smith_statement_final.pdf.

236. Greenberg, 946 N.Y.S.2d at 4.

237. United States v. Ferguson, 676 F.3d 260, 281 (2d Cir. 2011) (dealing

with key questions of who attended what meetings that were thus critical to

the question of conspiracy). On remand, the defendants entered into de-

ferred-prosecution agreements in which they acknowledged that “aspects of

the [General Re] transaction were fraudulent” and that they should have

“attempted to stop [the transaction] from going forward, but instead contin-

ued to participate in it.” Consent Motion for Deferred Prosecution Continu-

ance, United States v. Ferguson, No. 06-cr-137 (D. Conn. June 22, 2012);

Deferred Prosecution Order, United States v. Ferguson, No. 06-cr-137 (D.

Conn. June 25, 2012).

238. For the Wall Street Journal’s continued attack against New York’s case,

see, for example, The ‘Risk-Free’ Hank Greenberg Case,

ALL

. J. (

Feb. 19,

2015), http://www.wsj.com/articles/the-risk-free-hank-greenberg-case-1424

390839.

876 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

and under the Martin Act. The pressure to remove Greenberg

and Smith was undoubtedly heightened by the full array of

charges, many of which were subsequently withdrawn. One

possible interpretation of this maneuver is that the govern-

ment dropped all of its common-law counts because it did not

think that it could meet their distinctive scienter barrier. But

that explanation does not cover the decision to drop both com-

mon law and all but two Martin Act counts. Nor is it consistent

with the public admission of scienter in the two claims that

Greenberg and Smith settled. It seems likely that these other

claims were dropped for reasons that likewise apply to com-

mon-law and Martin Act claims, implicating the need to prove

false and material statements of fact. Otherwise, the prosecu-

tion would have no reason not to keep all of the Martin Act

counts.

The next point of interest is the resultant punishment,

and its integration with the earlier decision. Prior to settle-

ment, the question of remedy required special attention. Since

the litigation arose only after the General Re and CAPCO

transactions had been reversed, there was no need for an in-

junction once Greenberg and Smith were out of AIG. Simi-

larly, it would have been difficult for any private investors to

prove that they had lost money because they traded in reliance

on the false information allegedly generated by these transac-

tions, whose effect on the net earnings of AIG was uncertain.

To avoid both of these traps, the prosecution requested, as au-

thorized under the Martin Act amendments, equitable relief,

more specifically, “including but not limited to a ban on

[Greenberg’s and Smith’s] participation in the securities in-

dustry and a ban on serving as an officer or director of a public

company.”

239

The NYAG also sought disgorgement of cash bo-

nuses the defendants had received from AIG. In its second de-

cision on the case, the New York Court of Appeals held that

the award of disgorgement from Greenberg and Smith did not

depend on a showing of irreparable harm, but, after trial,

could be tailored to the overall facts and circumstances of the

case.

240

There is nothing exceptional about this approach: res-

titution for illicit gains is a long-established remedy the origins

239. See People v. Greenberg, 994 N.E.2d 838 (N.Y. 2013).

240. See People v. Greenberg, 54 N.E.3d 74 (N.Y. 2016).

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 877

of which lie at common law and not in modern courts of eq-

uity.

241

The harder question is whether the appropriate remedy,

if push had come to shove, would allow the NYAG to remove

Greenberg and Smith from their positions at AIG. The NYAG

has broad remedial discretion, but removal would be virtually

impossible to justify if Greenberg and Smith’s scienter were

not established. No strict-liability offense could justify such

draconian public intervention. Thus, once again, the issue of

scienter—although formally irrelevant under the Martin Act—

plays a role in the case. On this score, the initial question is

whether the removal of a person from private (that is, non-

governmental) office should be subject to a public official’s

oversight or reserved in the first instance to the discretion of

the organization’s governing body. There is of course no de-

bate that, to the extent that a particular statute requires that

an officer or director be barred from participation in the se-

curities industry, the decision is neither that of the public en-

forcer nor that of the board of directors. The matter has been

taken out of their hand by a clear statutory directive. Yet the

situation is quite different in connection with civil actions

where the finding of scienter still leaves open the choice of

remedy, which depends on a variety of factors dealing with fre-

quency, severity, and self-aggrandizement in the particular

case.

The governmental imposition of restitution, or even fines,

on individual defendants is perfectly appropriate to maintain

deterrence against particular actions. The great advantage of

using some combination of these two remedies is that they

only impose a cash transfer penalty on the defendants, without

imposing losses on innocent shareholders who played no part

in the decision. In addition, these targeted measures do not

distort the incentives or operation of the firm of which the

defendants had long been an integral part. But banning de-

fendants from work in a firm or an industry is necessarily far

more intrusive on firm activities, because CEOs and CFOs are

not fungible. Executive searches are so arduous and time-con-

suming precisely because the difference between the first- and

241. See Sinclair v. Brougham [1914] AC 398 (HL) 431 (“The case of a

chattel is easy: A shopkeeper delivers an article at the house of B in mistake

for the house of A. An action would lie against B for restitution.”).

878 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

second-best candidate could be the difference between success

and failure. Thus one key, if not the key, decision of any board

is the selection of its CEO, just as the key decisions of that

CEO involve the appointment of her chief subordinates.

In light of the critical relationship between executive se-

lection and firm success, the first-line decision on these mat-

ters to rest with the Board itself, which can of course take into

account the outcome of the civil prosecution under the Martin

Act and its proof of scienter. Clearly, care must be taken to

avoid conflicts of interest so that in this instance, neither

Greenberg nor Smith should participate in any of the delibera-

tions or vote on the outcome. If the Board, properly consti-

tuted, thinks, on the basis of its own internal deliberations,

that it should replace the defendants, that should be the end

of the matter, so long—and this is a critical qualification—as

its action was not taken under an explicit or implicit threat of

further investigations or litigation by any regulatory body, be it

the NYAG or the SEC. It is impossible for an outsider to pro-

ject how a board’s deliberations whether to fire an executive in

light of possible criminal sanctions will or should come out in

any particular case; in a sense, that is precisely the point. The

Board has better incentives than any insiders on the Board to

make the correct trade-offs. It will, of necessity, have more

complete knowledge of the circumstances surrounding the

particular sanctions, and it will be able to place any one deci-

sion in the larger context of the overall performance of the

corporate officers, like Greenberg and Smith, have provided

to the corporation through their entire tenure.

On this score, it surely should matter to the Board that

the historical success of the firm since Greenberg took over

had been legendary. The Board is likely also to find it in

Greenberg’s and Smith’s favor that they did not seek personal

enrichment, and neither of their actions had any direct effect

on the AIG’s bottom line in either the short or long run. Their

actions were undertaken to bolster the position of AIG in fi-

nancial markets, for it was undisputed that Greenberg and

Smith were deeply concerned about the underwriting profits

that they considered to be a bellwether test of the firm’s suc-

cess and failure. Accordingly, the transaction was intended to

change public perceptions of the firm’s position in the market-

place, and to increase, at least in the short run, the ability of

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 879

AIG to expand its business. A mistake in judgment, but one

subject to rectification.

In light of this record, if the Board, armed with full knowl-

edge, favors retention, it is difficult to come up with any strong

public-policy rationale for permitting the NYAG or the SEC to

overrule that decision. This intervention would impose upon

the firm the arduous duty to find a new management team

that would be responsible for overall firm direction. In retro-

spect, the decision to remove Greenberg and Smith may have

proved very costly because of the huge financial losses that

were incurred in the financial meltdown of 2008. To be sure, it

is always difficult to guess whether whether any CEO or CFO,

old or new, could have unwound AIG’s complex positions in

time to fix the problem. But the ousted pair’s large experience

in handling complex transactions might have led them to re-

duce AIG’s reliance on complex swaps, which put the firm at

risk during the crisis of 2008.

Rather than resorting to the ultimate remedy of dismissal,

a responsible board could have dealt with the matter by impos-

ing lesser sanctions: internal fines, additional oversight,

preclearance, and reporting requirements inside the firm; a

redefinition of responsibilities; a change in contracting prac-

tices; and a strong warning of what would happen if the error

were repeated, without forcing a precipitous change in top

management. If the entire system does not meet with public

approval, a decline in share value could offer a quick gauge of

public response to the initiative that could easily lead to a

change of firm direction. Indeed, in those cases where the

Board does not sufficiently respond to egregious CEO and

CFO behavior, shareholders have available private derivative

suits against the Board of Directors for breach of its duties of

care and loyalty. It is precisely because shareholders are not

likely to want to bring those disruptive actions that the entire

process of corporate governance should not be short-circuited

by officious, unilateral threats by regulatory authorities.

Spitzer’s serious lapse in judgment imposed threats against the

AIG board, which essentially forced their hand on a matter

that was better left to their judgment.

Once scienter is introduced into the case, it is of course

possible to pursue punitive action in these removal cases, but

no more so under the Martin Act than under the federal se-

curities laws, both of which go beyond the conventional com-

880 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

mon-law tort action’s limit to compensation for discrete past

harms. At this point, the strong equitable remedy of removal

from office would be questionable if it rested solely on bad

transaction outcomes without any proof of scienter, given the

huge sanctions requested and the inevitability of some transac-

tional failures under the best of circumstances. In complex

transactional cases, moreover, it should be clear that some sci-

enter requirement has to be read into the law, notwithstanding

the Martin Act’s apparent indifference to that element, be-

cause adopting some version of the business judgment rule is

the only way to cabin both civil and criminal liability within

reasonable bounds.

By way of background, publicly and privately traded com-

panies book all sorts of transactions in ways in which the ac-

counting methods do not perfectly track the underlying eco-

nomics of the situation. In some cases, the rules they follow are

dictated by either tax or securities authorities. In other cases,

they are the result of business judgments by the firm on how

best to account for certain complex transactions, given issues

such as the timing of income or deductions, or the allocation

of income or loss among various parties to myriad transac-

tions. It would impose an impossible regulatory burden on all

firms if any false public impression created by the choice of an

accounting method could expose a company to potential civil

or criminal liability without scienter.

The rule in practice, therefore, for both the federal secur-

ities acts and the Martin Act, has been that sanctions are im-

posed only where there is a known and deliberate flouting of

standard accounting rules of the sort found in the AIG case.

242

As to the other difficult transactions, the usual rule requires

firms to add notes to their financial statements to clarify cer-

tain balance sheet entries. Once that is done, the entire matter

is left to potential purchasers of these securities to ferret out

an accurate picture of the institutions’ financial health for

themselves. Ultimately, the question is who bears the risk of

inaccurate information. The accommodation under both the

securities acts and the Martin Act charges firms and their se-

nior officers with responsibility for sham or bad-faith transac-

tions, and nothing more. The rest of the risk rightly falls on

individual investors. The admissions of both Greenberg and

242. See supra notes 233–235 and accompanying text.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 881

Smith, like that of BMS, fall into the first class and thus break

no new ground.

It is also worth noting that the NYAG did not seek to im-

pose a deferred-prosecution agreement (DPA) on Greenberg,

Smith, or AIG. DPAs are commonly imposed under the federal

securities law, as illustrated in the federal prosecution in Fergu-

son.

243

The use of DPAs, however, has rightly given rise to seri-

ous condemnation because the warped prosecutorial incen-

tives they create can lead to systematic abuse.

244

Bringing a

lawsuit triggers huge collateral consequences in the form of

lost licenses to do business, which can be far more serious than

the penalties imposed upon conviction for the stated offense.

Conviction requires proof beyond a reasonable doubt in a

criminal case, yet the barriers to starting an investigation are

almost nil. Deciding to investigate a case requires no proof at

all, nor does a decision to file criminal charges. There is, ac-

cordingly, less protection against this serious charging practice

than there is against conviction, and this structural defect al-

lows prosecutors to push to get extensive settlements. Just this

dynamic happened under the federal securities laws when J.P.

Morgan Chase was bludgeoned into a $1.7 billion fine for its

supposed role in not detecting the Bernard Madoff scandal,

even though J.P. Morgan was not mentioned even once in the

SEC’s own exhaustive report on how the scandal had gone un-

detected for so long. That report first noticed that the SEC

had all sorts of “red flags” in dealing with Madoff, but none-

theless managed to exonerate all SEC employees of “any finan-

cial or other inappropriate connection with Bernard Madoff

or the Madoff family that influenced the conduct of their ex-

amination or investigatory work.”

245

Nonetheless, the SEC

243. See Copland & Mangual, supra note 34.

244. See, e.g., James Copland & Isaac Gorodetski, The Shadow Lengthens: The

Continuing Threat of Regulation by Prosecution

, M

ANHATTAN

NST

. (

Feb. 25,

2015), https://www.manhattan-institute.org/html/shadow-lengthens-contin

uing-threat-regulation-prosecution-5898.html; Epstein, The Dangerous Incen-

tive Structures, supra note 34; Epstein, Deferred Prosecution Agreements on Trial,

supra note 30.

245.

U.S. S

. & E

XCH

. C

OMM

’

, O

NVESTIGATIONS

, I

NVESTIGATION

AILURE OF THE

SEC

NCOVER

ERNARD

ADOFF

’

ONZI

CHEME

—

UBLIC

ERSION

(2009), http://www.sec.gov/news/studies/2009/oig-509.

pdf. For my critique, see Richard A. Epstein, In Defense of JP Morgan,

OOVER

NST

.: D

EFINING

DEAS

(Oct. 29, 2013), http://www.hoover.org/research/de-

fense-jp-morgan. For the evolution of the case, see Christian Dem, JPMorgan

882 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

used its prosecutorial power against J.P. Morgan, whose sole

responsibility was to make routine disbursements to Madoff’s

clients, from which it did not and could not infer any illegality.

Notably, those heavy-handed tactics were not employed

against Greenberg or Smith. On the financial side, the only

remedy was disgorgement of bonuses received during the

2001–2004 period when these transactions were still on the

books. This was accompanied by an allowance for the dis-

gorgement previously made to the federal government for the

case brought and settled under the federal securities laws. This

complementarity shows that the common elements between

the two statutes were greater than any differences. There is lit-

tle reason to think that the major dispute over the AIG transac-

tions will result in unanimous agreement. But for these pur-

poses at least, the potential dangers that are found under the

Martin Act are also present under the federal securities laws,

which give rise to dangers of dual enforcement and an abuse

of the investigative process. But in this case at least, the admis-

sion on scienter dampens the charges with respect to liability,

and the admitted liability therefore leaves us uncertain as to

whether a court would have removed Greenberg and Smith

had they not resigned under pressure.

C. Global Warming: Fraud and Free Speech

1. The Litigation

a. ExxonMobil

The most recent Martin Act situation involves the cam-

paign started by NYAG Schneiderman to invoke the Martin

Act to attack ExxonMobil for allegedly creating a false climate

of public opinion that would allow the company to artificially

raise the price of its stock. The supposed technique consisted

of a decades-long effort by ExxonMobil to poo-poo the science

on global warming, in order to dampen the incredible legal

and social pressure to curtail or eliminate the use of fossil fuels

that lie at the heart of its business. The simple argument put

Chase Could Face Penalties Related to Madoff Scandal,

AILY

(Oct. 24, 2013),

http://www.dailykos.com/story/2013/10/24/1250190/-JPMorgan-Chase-

could-face-penalties-related-to-Madoff-scandal#; see also Jessica Silver-Green-

berg & Ben Protess, JPMorgan Chase Nears a $2 Billion Deal in a Case Tied to

Madoff,

N.Y. T

IMES

(Jan. 5, 2014, 10:00 PM), https://dealbook.nytimes.com/

2014/01/05/jpmorgan-chase-nears-a-2-billion-deal-in-a-case-tiedo-madoff/.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 883

forward by Schneiderman was that the First Amendment does

not protect against fraud, and a pervasive industry fraud is

what ExxonMobil perpetuated. The baldness of his claim was

made clear in his March 29, 2016 press conference that an-

nounced the legal assault, brought in part under the Martin

Act

246

[W]e are here for a very simple reason. We have

heard the scientists. We know what’s happening to

the planet. There is no dispute but there is confu-

sion, and confusion sowed by those with an interest

in profiting from the confusion and creating mis-

perceptions in the eyes of the American public that

really need to be cleared up. The U.S. Defense De-

partment, no radical agency, recently called climate

change an urgent and growing threat to our national

security. We know that last month, February, was the

furthest above normal for any month in history since

1880 when they started keeping meteorological

records. The facts are evident. This is not a problem

ten years or twenty years in the future. . . .

[ExxonMobil is] using the best climate models so

that when they spend shareholder dollars to raise

their oilrigs, which they are doing, they know how

fast the sea level is rising. Then they are drilling in

places in the Arctic where they couldn’t drill 20 years

ago because of the ice sheets. They know how fast the

ice sheets are receding. And yet they have told the

public for years that there were no “competent mod-

els,” was the specific term used by an Exxon executive

not so long ago, no competent models to project cli-

mate patterns, including those in the Arctic. And we

know that they paid millions of dollars to support or-

ganizations that put out propaganda denying that we

can predict or measure the effects of fossil fuel on

246. For this transcript, see Appendix in Support of Petition and Emer-

gency Motion of Exxon Mobil Co. to Set Aside or Modify the Civil Investiga-

tive Demand or Issue a Protective Order at App. 002–021, In re Civil Investi-

gative Demand No. 2016-EPD-36 (Mass. Sup. Ct. June 10, 2016) (No. 16-

1888F) [hereinafter Massachusetts Superior Court Appendix], http://

www.mass.gov/ago/docs/energy-utilities/exxon/appendix-in-support-of-pe

tition-and-emergency-motion.pdf.

884 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

our climate, or even denying that climate change was

happening.

There have been those who have raised the question:

aren’t you interfering with people’s First Amendment

rights? The First Amendment, ladies and gentlemen,

does not give you the right to commit fraud. And we

are law enforcement officers, all of us do work, every

attorney general does work on fraud cases. And we

are pursuing this as we would any other fraud matter.

You have to tell the truth. You can’t make misrepre-

sentations of the kinds we’ve seen here.

247

These statements are not unique to Schneiderman, for

similar findings were part of a proposed California statute, the

Climate Science Truth and Accountability Act of 2016, which

would have allowed similar climate-related fraud prosecutions

under the California unfair competition and consumer protec-

tion laws

248

Sec. 2. (a) The legislature finds and declares all of

the following:

There is broad scientific consensus that anthropo-

genic global warming is occurring and changing the

world’s climate patterns, and that the primary cause

is emission of greenhouse gases from the production

and combustion of fossil fuels, such as coal, oil and

natural gas.

249

Outside the Martin Act, it is possible to find literally hun-

dreds of statements that make the same proposition: the sci-

ence on global warming is clear, and it is settled, so it is only

the obstructionists and the deniers who block the path to so-

cial redemption. Thus, a group of prominent scientists have

written: “Human-caused climate change is not a belief, a hoax,

or a conspiracy. It is a physical reality. Fossil fuels powered the

Industrial Revolution. But the burning of oil, coal, and gas also

caused most of the historical increase in atmospheric levels of

heat-trapping greenhouse gases. This increase in greenhouse

247. Id. at App. 003–005.

248. S.B. 1161, 2015–2016 Reg. Sess. (Cal. 2016), https://leginfo.legisla

ture.ca.gov/faces/billCompareClient.xhtml?bill_id=201520160SB1161.

249. Id.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 885

gases is changing Earth’s climate.”

250

Note that this letter con-

tains no specifics, but does assert that climate change is re-

sponsible for “sea level rise, altered rainfall patterns, retreat of

Arctic sea ice, ocean acidification, and many other aspects of

the climate system.”

251

More specifically, global warming has been advanced to

explain the recent break off of a huge, trillion metric ton ice-

berg from the Larsen C ice shelf in Antarctica. According to

David Remnick of the New Yorker, “[s]uch events now seem al-

most ordinary—and harbingers of far worse.”

252

It is quite pos-

sible that, “should the much larger West Antarctic Ice Sheet

thaw and slip into the ocean, sea levels across the globe could

rise as much as seventeen feet.”

253

In other writings, sober-

minded writers working through the Climate Leadership

Council,

254

including George Shultz, Jim Baker, and Martin

Feldstein, have taken the position that “[m]ounting evidence

of climate change is growing too strong to ignore. While the

extent to which climate change is due to man-made causes can

be questioned, the risks associated with future warming are too

big and should be hedged. At least we need an insurance pol-

icy.”

255

Their recommendation was that the United States impose

a $40 per ton tax on carbon dioxide (CO

) emissions, subject

to gradual increase, which will be then returned to the public

in the form of a carbon dividend that they hope will exceed

any additional energy costs that most Americans would have to

bear. The tax will be protected against international arbi-

trage—i.e., selling goods made outside the US by companies

250. An Open Letter Regarding Climate Change From Concerned Members of the

U.S. National Academy of Sciences,

ESPONSIBLE

CIENTISTS

(Sept. 20, 2016),

http://www.responsiblescientists.org.

251. Id.

252. David Remnick, Trump Family Values,

ORKER

(July 24, 2017),

http://www.newyorker.com/magazine/2017/07/24/trump-family-values.

253. Id.

254. For a description of the organization, see Founding Statement,

LIMATE

EADERSHIP

OUNCIL

(June 20, 2017), https://www.clcouncil.org/founding-

statement/.

255.

AMES

A. B

AKER

, III, M

ARTIN

ELDSTEIN

, T

ALSTEAD

, N. G

REGORY

ANKIW

, H

ENRY

M. P

AULSON

, J

. G

EORGE

P. S

HULTZ

, T

HOMAS

TEPHENSON

ALTON

LIMATE

EADERSHIP

OUNCIL

ONSERVATIVE

ASE FOR

ARBON

IVIDENDS

1 (2017), https://www.clcouncil.org/media/TheConser

vativeCaseforCarbonDividends.pdf.

886 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

that do not have to pay local carbon taxes—by imposing bor-

der carbon adjustments. These three elements in turn will al-

low for the removal or rollback of direct forms of regulation,

including the Obama Clean Power Plan, and would also blunt

any effort to use the tort system to impose liability on separate

companies—which has long been regarded as a most unrealis-

tic proposal given the impossible logistics of litigation.

256

Still

other writers, most notably David Wallace-Wells, predict more

dire consequences: an “Uninhabitable Earth,” from which we

can expect “[f]amine, economic collapse, [and] a sun that

cooks us.”

257

There is little doubt that these are dire statements. But

wholly apart from the question of their accuracy, there is the

difficult question about the connection to the Martin Act it-

self. The first point to note is that the definition of fraud used

by Schneiderman in his press conference is one of hardcore

legal fraud, which posits that ExxonMobil knew all the rele-

vant environmental impact information that is taken, without

demonstration, as though it were a key fact. If his account were

correct, he would have no need to rely on the expanded defi-

nition of fraud that dispenses with scienter, which in this con-

text does raise serious First Amendment issues, given the sup-

pression of debate that could follow if all statements of opin-

ion that proved false were actionable. Indeed, an SEC

composed of five Schneidermans could bring exactly the same

case under the federal securities laws, using the same narrow

definition of scienter. In both cases, any claim that Exx-

onMobil should be subject to sanction even if it were simply

unaware of the truth when it issued its reports and made its

statements would be dead-on-arrival as an approach to stimu-

late public outrage against the company. If Exxon is the only

firm that did serious climate research, it would be ironic that it

256. For a rejection of this theory, at least for federal courts, see Am. Elec.

Power Co. v. Connecticut, 564 U.S. 410 (2011). The proposal has been sup-

ported by Benjamin Ewing & Douglas A. Kysar, Prods and Pleas: Limited Gov-

ernment in an Era of Unlimited Harm, 121

ALE

L.J.

350 (2011). For my re-

sponse, see Richard A. Epstein, Beware of Prods and Pleas: A Defense of the Con-

ventional Views on Tort and Administrative Law in the Context of Global Warming,

121

ALE

L.J. O

NLINE

317 (2011).

257. David Wallace-Wells, The Uninhabitable Earth,

N.Y. M

(July 9, 2017),

http://nymag.com/daily/intelligencer/2017/07/climate-change-earth-too-

hot-for-humans.html.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 887

alone should be held responsible for its foresight. But if other

oil companies did the same, why exempt them from liability

given that they could also be subject to suit under the current

scienter theory?

The popular campaigns against ExxonMobil take the

same form. None of these efforts rely on anything that re-

motely resembles a strict-liability theory under which proof of

the defendant’s negligence or intention to harm is irrelevant.

Schneiderman was well aware that this theory could not sup-

port any kind of financial remedy or injunctive relief against

ExxonMobil, without imposing potential liability on virtually

everyone else who had made some erroneous climate forecast.

The most that could be sought would be some requirement,

similar to that imposed on AIG, for misstating reserves or

other metrics used in the oil and gas business. Far from ap-

pealing to strict liability, the hashtag used by opponents of Ex-

xon was, and remains, #ExxonKnew.

258

These opponents do

not rely on the niceties of the Martin Act when they proclaim:

“Exxon knew about climate change half a century ago” and

“[t]hey deceived the public, misled their shareholders, and

robbed humanity of a generation’s worth of time to reverse

climate change.”

259

Each of these references speaks in terms of

deception, duplicity, and worse.

One notable expos

e in the Los Angeles Times, “What Exxon

Knew About the Earth’s Melting Arctic,”

260

claims Exxon

downplayed the risks of global warming externally as its own

scientists were “closely studying the impact of climate change

on the company’s operations.”

261

The article noted that the

change in Arctic conditions was both good and bad news for

the company. On the one hand, it reduced the cost of explora-

tion in the oil-rich Arctic; on the other hand, climate change

258. See

XXON

NEW

, http://exxonknew.org/ (last visited Nov. 19,

2017).

259. Id. See also Neela Banerjee et al., CO

’s Role in Global Warming Has Been

on the Oil Industry’s Radar Since the 1960s,

NSIDE

LIMATE

EWS

(Apr. 13,

2016), https://insideclimatenews.org/news/13042016/climate-change-

global-warming-oil-industry-radar-1960s-exxon-api-co2-fossil-fuels;

AOMI

RESKES

& E

RIK

ONWAY

ERCHANTS OF

OUBT

: H

OW A

ANDFUL OF

CIEN-

TIST

BSCURED THE

RUTH OF

SSUES FROM

OBACCO

MOKE TO

LOBAL

ARMING

(2010).

260. Sara Jerving et al., What Exxon Knew About the Earth’s Melting Arctic,

L.A. T

IMES

(Oct. 9, 2015), http://graphics.latimes.com/exxon-arctic/.

261. Id.

888 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

could create “higher sea levels and bigger waves,” and poten-

tially impede drilling operations. The claim was that Exxon

wanted to tamp down claims of global warming to reduce pres-

sure to keep fossil fuels in the ground. But at no point does

the article explain in what way Exxon had or relied on any

private information of value when the standard temperature

measures were at all times in the public domain. Nor did it

bother to mention that, at the time the Los Angeles Times wrote

its article, the level of sea ice in the Arctic was in the process of

recovery from its 2012 low, which was inconsistent with their

doomsday view of the underlying science.

262

The denunciations, moreover, quickly reached a fever

pitch. One representative of this style of argumentation is a

report entitled “The Climate Deception Dossiers: Internal Fos-

sil Fuel Industry Memos Reveal Decades of Corporate Dis-

information.”

263

The expos

e starts with the observation that

half of all carbon emissions have taken place since 1988

264

—

without noting the weak correlation between those higher

emissions and the steady global temperatures over the same

period.

265

It also claims that “[g]lobal warming is accelerating

the rate of sea level rise and dramatically increasing flooding

risks,”

266

without addressing any of the abundant evidence to

the contrary. The study notes that some of these companies

“continue to support groups that spread misinformation de-

signed to deceive the public about climate science and climate

policy,”

267

again without explaining why the work of such

groups as the CO

Coalition is flawed, resting as it does on its

interpretation of independently generated and publicly availa-

ble data. Indeed, this critique does not systematically engage

the views of any scientists on the other side of the debate, let

alone address any of the particular points raised above. In-

stead, the constant trope is that the “deniers,” a clear allusion

262. Id.

263. See

ATHY

ULVEY

& S

ETH

HULMAN

NION OF

ONCERNED

CIEN-

TISTS

LIMATE

ECEPTION

OSSIERS

: I

NTERNAL

OSSIL

UEL

NDUSTRY

EMOS

EVEAL

ECADES OF

ORPORATE

ISINFORMATION

(2015), http://

www.ucsusa.org/sites/default/files/attach/2015/07/The-Climate-Decep

tion-Dossiers.pdf.

264. Id. at 3.

265. See infra text accompanying notes 320–322.

266.

ULVEY

& S

HULMAN

supra note 263, at 2.

267. Id.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 889

to Holocaust deniers, practice nonstop deceit in the effort to

preserve their business.

Here is one final example. Inside Climate News published

a series of expos

es on the subject in the last half of 2015, which

allowed no room for doubt. Some of the tenor comes through

in the headlines of the separate studies: “Exxon’s Own Re-

search Confirmed Fossil Fuels’ Role in Global Warming De-

cades Ago: Top executives were warned of possible catastro-

phe from greenhouse effect, then led efforts to block solu-

tions.”

268

To give some sense as to the overstatement in these stud-

ies, it is useful to look at just one of the documents that Inside

Climate News treats as a smoking gun. That document was a

June 6, 1978 presentation by J.F. Black, a scientific advisor at

Exxon, to high-ranking Exxon executives.

269

That presenta-

tion refers to an earlier oral presentation, the graphs from

which are included at the end of the report, that he had made

to the Exxon Corporate Management Committee the previous

July. It is exceedingly difficult to distill the ultimate message

from that report, given its constant diffidence on the sound-

ness of the underlying theory or the available data. Thus it

notes “[t]he CO

increase measured to date is not capable of

producing an effect large enough to be distinguished from

normal climate variations.”

270

Shortly thereafter his report

states: “There is considerable uncertainty regarding what con-

trols the exchange of atmosphere CO

with the ocean and with

carbonaceous materials on the continents,” and, “[m]odels

which predict the climatic effects of a CO

increase are in a

primitive stage of development.”

271

Black’s report is replete

with further references to the uncertainties in question. In one

telling section, Black discusses the effect of the “doubling of

carbon dioxide concentrations in the atmosphere,” and con-

268. Neela Banerjee, Lisa Song & David Hasemyer, Exxon’s Own Research

Confirmed Fossil Fuels’ Role in Global Warming Decades Ago,

NSIDE

LIMATE

EWS

(Sept. 16, 2015), http://insideclimatenews.org/news/15092015/Exx

ons-own-research-confirmed-fossil-fuels-role-in-global-warming.

269. J.F. Black, Sci. Advisor, Exxon Res. & Eng’g Co. Prods. Res. Div., The

Greenhouse Effect: Transcript of a Talk Delivered Before the PERCC Meet-

ing (May 18, 1978), http://insideclimatenews.org/sites/default/files/docu

ments/James%20Black%201977%20Presentation.pdf.

270. Id. at 1.

271. Id. at 2.

890 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

cludes “this will have a negligible effect on sea levels” and

“probably no effect on the polar ice sheets.”

272

At another

place, his report says:

A study of past climates suggests that if the earth does

become warmer more rainfall should result. But an

increase as large as 2

C would probably also affect the

distribution of the rainfall. A possible result might be

a shift of both the desert and the fertile areas of the

globe toward higher latitudes. Some countries would

benefit but others could have their agricultural out-

put reduced or destroyed. The picture is too unclear

to predict which countries might be affected favora-

bly or unfavorably.

273

The Inside Climate News report quotes only one sentence

from this passage: “Some countries would benefit but others

could have their agricultural output reduced or destroyed.”

274

The lead-in to this quotation reads: “He warned Exxon scien-

tists and managers that independent researchers estimated a

doubling of the carbon dioxide (CO

) concentration in the

atmosphere would increase average global temperatures by 2

to 3 degrees Celsius (4 to 5 degrees Fahrenheit), and as much

as 10 degrees Celsius (18 degrees Fahrenheit) at the poles.”

275

Black’s letter does not use the word “warning,” and it also con-

tains this sentence: “More research is needed, however, to es-

tablish the validity and significance of predictions with respect

to the Greenhouse effect.”

276

This passage was presented in the final “Summary” of the

report, but it offers no evidence for its most striking proposi-

tion—that any increase in the earth’s temperature will gener-

ate a three-fold increase at the poles. The basic theory at most

explains why some greater polar increase should occur, but

gives no clue for what that number or effect would be. In addi-

tion, previous temperature changes, either up or down, did

not show such an extreme reaction at the poles. Black’s letter

does not give any indication whether other scientists inside the

company accepted his presentation. Indeed, the last sentence,

272. Id. at 7.

273. Id. at 1.

274. Id.

275. Banerjee, Song & Hasemyer, supra note 268.

276. Black, supra note 269, at 10.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 891

which was an obvious call for further research, is inconsistent

with the tenor of the rest of the report. And no one has of-

fered any reason to think that there was a groundswell of pro-

fessional opinion, inside or outside Exxon, agreeing with this

prediction, which was after all, based largely on publicly availa-

ble information.

It should go without saying that virtually all of the Exxon

documents now available for public viewing are fair game for

investigative journalists determined to cherry pick data in or-

der to make a point. Yet ironically, the moderate level of the

observed changes in global temperatures are perfectly consis-

tent with the subsequent uncertainty expressed by Exxon on

the impact of global warming. This insistent pillorying of Exx-

onMobil does not make sense if the critics themselves have, as

seems to be the case, overstated their position on the impact of

global warming. Yet Inside Climate News never noted that

Black’s dire prediction of 10

C changes in polar temperature

had proved resoundingly false, given that the amount of Arctic

sea ice had already rebounded from its 2012 low when its story

came out in 2015. Indeed, the entire condemnation rests on

the unexamined assumption that the harmful effects of global

warming have been conclusively established. Such is the in-

complete and inaccurate state of the evidence on which

Schneiderman relied.

It is, of course, possible to amass evidence that under-

mines any skeptical account of global warming. But for the

purposes of the Martin Act, the ultimate truth on the scientific

question is not what matters. For these purposes, the key issue

is whether the categorical statements made about the defini-

tive science on global warming are defensible when made by

the NYAG, without bothering to look at any of the theoretical

and empirical evidence lined up on the other side of the de-

bate. For the purposes of the Martin Act, it does not matter

that lots of people, including many reputable scientists, agree

with Schneiderman. What matters is whether any supposed

consensus view on a matter of scientific concern should be al-

lowed to shut down scientific debate. I have gone through a

fraction of the evidence with some care in order to make clear

that the proper object of public condemnation is less the Mar-

tin Act than the dogmatic insistence by Schneiderman and

others that these tough issues of climatology are so cut-and-

dried that only one view of them is acceptable in public dis-

892 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

course, and ultimately definitive of the boundaries of legal

fraud.

In response to the pointed criticism, ExxonMobil has

taken two lines of defense. The first is to announce, along with

many other oil companies, its support of the carbon tax propo-

sal advanced by Baker and Shultz.

277

Their support has been

mocked as skin deep by environmental groups such as 350.org

(which claims that any level of CO

that exceeds 350 number

of parts per million is dangerous, including the current levels

that are now over 400 ppm).

278

And it is always difficult to fig-

ure out dual-motive cases when it is clear that ExxonMobil

fought back with uncommon vigor against the requests of

Schneiderman to continue its investigation into its accounting

practices. At this point in time, the entire unfortunate matter

is bogged down in the details of pre-trial motions. ExxonMobil

pleaded in vain with New York federal district judge Valerie E.

Caproni, to take depositions of Schneiderman and Massachu-

setts Attorney General Maura Healey in an effort to establish

their abuse of power.

279

At the same time, in New York State

Court, ExxonMobil attacked Schneiderman’s “frivolous”

claims, which it insists are filled “with inflammatory, reckless

and false allegations of an ‘ongoing fraudulent scheme’ and

‘sham’ business practices.”

280

They added, “No further evi-

dence is required to establish the political motivation of the

attorney general’s fruitless year-and-a-half long investigation

pursuing his ever-shifting and unraveling investigative theo-

ries.”

281

The first of these theories relates to the knowledge

277. John Schwartz, Exxon Mobil Lends Its Support to a Carbon Tax Proposal,

N.Y. T

IMES

(June 20, 2017), http://www.nytimes.com/2017/06/20/science/

exxon-carbon-tax.html.

278. Id.

279. David Hasemyer & Nicholas Kusnetz, Exxon’s Opposition to Climate

Fraud Probes Gets Less Sympathy from NY Judge,

NSIDE

LIMATE

EWS

(

Apr. 21,

2017), https://insideclimatenews.org/news/21042017/exxon-mobil-cli

mate-change-scandal-fraud-investigations-new-york-schneiderman.

280. Erik Larson, Exxon Says N.Y. Attorney General Distorted Company’s Cli-

mate Data,

LOOMBERG

(June 9, 2017, 4:12 PM), https://www.bloomberg

.com/news/articles/2017-06-09/exxon-says-n-y-attorney-general-distorts-

company-s-climate-data.

281. John Siciliano, Exxon Mobil Blasts New York AG’s ‘Sham’ Climate Investi-

gation,

ASH

. E

XAM

’

(June 9, 2017), http://www.washingtonexaminer.

com/exxon-mobil-blasts-new-york-ags-sham-climate-investigation/article/26

25545.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 893

that Exxon (before it was ExxonMobil) had forty years ago.

The second is a claim that ExxonMobil today has manipulated

the evidence on the costs of regulatory compliance that should

be expected in the light of anticipated climate control regula-

tions.

The record on the historical claims is far more favorable

to Exxon than Schneiderman’s allegations suggested. Anyone

who cares to look at the reams of information already in the

public sphere would see that much of the research undertaken

by ExxonMobil on the subject largely consisted of its own in-

terpretation of publicly available studies. The Martin Act case

against Exxon bore absolutely no relationship to the flagrantly

false statements of private information contained in a business

prospectus, targeted to a particular vulnerable population,

which provoked the strong endorsement of the Martin Act

back in 1926 in Federated Radio Corp.

Note the differences in the two scenarios. With global

warming, it is impossible to link the research to any effort to

influence the buying or selling of shares at all. The statements

were made at odd times, and were never associated with the

purchase or sale of any security or with any effort to buy or sell

any securities, both of which are explicit limitations on the

NYAG’s authority under the Martin Act. In addition, given the

wealth of information that forms the backdrop of all of Ex-

xon’s work, it seems clear that none of the asserted falsehoods

would count as a false statement of fact, to which the Martin

Act is restricted. These clearly would fall under the class of

predictions, as to the future course of world climate change,

or opinions about the same course of action. No person could

possibly or properly rely on this information to make a deci-

sion to purchase shares, which means that, given the context,

anyone would be hard pressed to say that the information pro-

duced by Exxon was material in the sense that this term is used

in the securities law (as Columbia Law Professor Merritt Fox

has noted in a public panel on the topic

282

). As discussed ear-

lier, there is a proper distribution of responsibility for error in

cases of this sort. The defendant companies are only held re-

282. See Chris White, Law Expert Blasts NY Attorney General for Investigations

Against Exxon,

AILY

ALLER

(June 7, 2016, 4:22 PM), http://dailycaller.

com/2016/06/07/law-expert-blasts-ny-attorney-general-for-investigations-a

gainst-exxon/.

894 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

sponsible for what are known to be sham transactions. For the

rest, it is the duty of all investors to seek out information on

their own to determine the extent, sources and effect of global

warming on the use of fossil fuels. They are free to think that

the prospects are dire, or that the entire matter may blow over.

The legal implications seem clear. No decided case of the

Martin Act has ever applied the term “fraud” to transactions

that are wholly unrelated to the buying and selling or the po-

tential buying and selling of stock. Whether one accepts the

narrower definition of “in connection with” that was used in

Blue Chip Stamps to exclude statements which were not fol-

lowed by a purchase or sale, or the broader views under the

Martin Act, which allow for the injunctions of potential sales,

the answer is the same. These statements have so many pur-

poses that they could not fall into any class of false statement

of facts. There are, of course, First Amendment issues in these

cases, and that body of law does not offer any protection for

false statements of facts made in pursuit of a sale. But it offers

extensive protection for statements of opinion based on mat-

ters of public interest and concern, in which discourse regard-

ing the future outcome of scientific events would surely fall.

283

Ultimately, Schneiderman was right that the First Amendment

does not protect against fraud. But he is wrong to think that

this case comes close to that general prescription. None of the

proponents of serious global warming has a monopoly on the

truth. As Oliver Wendell Holmes said a long time ago in

Abrams v. United States, “the best test of truth is the power of

the thought to get itself accepted in the competition of the

market,”

284

which means that no one gets a pass against criti-

cism.

The second Schneiderman charge switched from histori-

cal knowledge to the claim that ExxonMobil used different

numbers for internal and external consumption in seeking to

measure the costs, if any, of potential direct government taxes

or regulation on CO

emissions. This claim has a more con-

temporary focus, but it is subject to the same general objec-

283. See Gertz v. Robert Welch, Inc., 418 U.S. 323 (1974) (strengthening

First Amendment protection for speech dealing with a matter of public con-

cern); Dun & Bradstreet, Inc. v. Greenmoss Builders, Inc., 472 U.S. 749

(1985) (refusing to extend like protection to statements that do not have

public interest and concern).

284. Abrams v. United States, 250 U.S. 616, 630 (1919).

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 895

tions. It is, of course, common to use different numbers for

different purposes and in any event, even under the Martin

Act, the proof of materiality would require some judgment

that ordinary investors would be influenced by this one fact,

among many, in their decision to buy stock.

Inquiries of this sort are unlikely to bear fruit for two rea-

sons. First, there are many forms of public information availa-

ble regarding the prospect of government regulation or taxa-

tion—investors could look to independent sources to make

their own judgment as to the probability and severity of antici-

pated government regulation in the United States or else-

where. They need not rely on information provided by Exx-

onMobil. As Professor Fox put the point in 2016: “the market

was well supplied with information about climate change.”

285

It also matters that the impact of any such regulation lies far

into the future, and thus must be discounted down to present

value to calculate its impact, if any, on share prices. As such,

the business judgments of ExxonMobil bear no resemblance

to the discrete sham transactions with immediate financial im-

pact that led to the favorable Martin Act settlements against

AIG, Greenberg, and Smith.

Second, a great number of issues influence movements in

stock price for large publicly traded companies, so it is highly

unlikely that projections about the cost of regulations that may

be forthcoming ten or twenty years into the future would be

accorded much weight by investors who know how rapidly the

political landscape can move. The election of Donald Trump

as president, the appointment of Scott Pruitt as head of the

Environmental Protection Agency, and the decision to with-

draw from the Paris Accords show just how quickly events can

overtake predictions. In addition, the overall success of Ameri-

can firms in reducing their CO

footprint without further reg-

ulation cuts in the same direction. Any substantial reduction in

emissions largely depends on what China and India do in

the next decade, again, independent variables for individuals

to weigh on their own. It is hard to point to any uniquely pri-

vate information that ExxonMobil had that could influence

share prices. Schneiderman’s campaign has all the looks of a

vendetta.

285. White, supra note 282.

896 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

This analysis has been borne out by subsequent events. In

February 2017, ExxonMobil’s annual 10-K filing announced

that the company had cut its estimate of recoverable reserves

by about 3.3 billion barrels, out of a total of just more than 23

billion barrels.

286

The amount of barrels cut was not chosen at

random, but represented the oil found in the costly to extract

Canadian oil sands. The driver was not the prospect of future

regulation intended to cope with global warming, but the

more prosaic reality that at present the oil contained in these

sands was too costly to extract economically under the prevail-

ing low prices—the most material of all circumstances. The

move was widely anticipated and, as even the pro-global warm-

ing website, Inside Climate News acknowledged, had nothing

to do with climate change.

287

The lessons here are two-fold.

First, the market does indeed know how to value stocks, ren-

dering global warming regulation an exotic irrelevancy. Sec-

ond, there are no substantive differences between the federal

securities laws and the Martin Act. Both the SEC and the

NYAG could investigate the changes, and both should back off

in light of the changes that did occur. Let this be a lesson for

future cases.

b. Global Warming and Public Nuisance

The question then becomes whether this initial foray

against ExxonMobil contains any lessons for dealing with the

more recent set of suits that have been filed by a number of

municipalities in New York,

288

California and elsewhere on the

question of whether global warming constitutes a public nui-

sance that has been created by ExxonMobil and other oil com-

286. Geoffrey Smith, Exxon’s Big Oil Sands Write-Off Could Help It Dodge SEC

Troubles,

ORTUNE

(Feb. 23, 2017), http://fortune.com/2017/02/23/exxon-

mobil-oil-sands-sec/.

287. Nicholas Kusnetz, Exxon Relents, Wipes Oil Sands Reserves from Its Books,

NSIDE

LIMATE

EWS

(Feb. 23, 2017), https://insideclimatenews.org/news/

22022017/exxon-mobil-tar-sands-alberta-canada-climate-change-oil-prices

(“Low prices force the oil giant to stop counting 3.5 billion barrels as an

asset, a major turnaround in its accounting.”).

288. Complaint, New York v. BP PLC (S.D.N.Y. 2018) (No. 18-cv-182)

(footnote omitted), http://www1.nyc.gov/assets/home/downloads/pdf/

press-releases/2018/complaint-filed-8031957-20180109.pdf.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 897

panies, which make and distribute fossil fuels throughout the

United States.

289

Much abbreviated, the key allegations in the New York

complaint read:

Defendants’ past and ongoing actions are harming

New York City now: the City already has suffered

damage from climate change, including inundation,

erosion, and regular tidal flooding of its property.

The City now faces further imminent threats to its

property, its infrastructure, and the health and safety

of its residents. . . .

Global warming-induced sea level rise is expected to

be higher in areas surrounding New York City than in

many other parts of the world. According to the

NPCC’s [New York City’s expert committee] “high

end” projection, the sea level surrounding the City is

expected to rise above the 2000-2004 baseline level

(which already includes climate-change related sea

level rise) by ten inches by the 2020s, by thirty inches

by the 2050s, by fifty-eight inches by the 2080s, and by

seventy-five inches—more than six feet—by 2100.

Even the “middle range” projections are dire: four to

eight inches by the 2020s, eleven to twenty-one in-

ches by the 2050s, eighteen to thirty-nine inches by

the 2080s, and twenty-two to fifty inches by 2100.

Even without storms, this sea level rise threatens low-

lying areas of the City—for example, by the 2050s ap-

proximately forty-three miles of the City’s coastline

(including many residential neighborhoods) could

be at risk of daily or weekly tidal inundation, even

during non-storm conditions. And a sea level rise of

six feet would put parts of all five boroughs —includ-

ing portions of the Financial District, Red Hook, and

the vast majority of Coney Island and the Rocka-

ways—under water.

290

289. For the record, I have worked as an advisor to the Manufacturer’s

Accountability Project on the legal issues that have stemmed out of these

lawsuits.

290. Complaint, supra note 288, at ¶2.

898 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

The public nuisance claim here is novel in all its as-

pects.

291

Previously in the American Electric Power Co. v. Connecti-

cut, the Supreme Court held that this cause of action was not

viable under federal law but left open the prospect that it

might be brought under state law, at least if the doctrines of

federal preemption did not block the case.

292

But this public

nuisance case is markedly different from AEP because it is not

brought against the emitters of greenhouse gases, but the pro-

ducers of fossil fuels for making and distributing these prod-

ucts.

293

The extra layer creates a major question of causation,

given that the total amount of carbon dioxide emissions is de-

pendent heavily on the way in which the crude oil is processed

and used by downstream parties. For these purposes, I shall

not discuss some of the nuisance law aspects of the case,

294

except to note that virtually all of these claims take the posi-

tion that in light of global warming, the risk of sea level rise is

far greater than it has ever been before so that it is appropriate

to require these defendants to fund an abatement program

that can stave off disaster before it arrives. One key difficulty

with this claim is that it is uncertain how the rate of carbon

dioxide increases translates itself into sea level rise. In fact, his-

torical averages have been roughly in the order of 1/20th of

an inch per year or five inches per century—a change that is

not obviously driven by carbon dioxide increases,

295

and that

rate has been constant over the past 20 years.

For these purposes, however, one question that arises

under both the Martin Act and the federal securities law is

whether the stark claims that are raised in these complaints

are inconsistent with the representations that most local gov-

ernments have made in their various bond offerings that take

the position that denies that there are any serious environmen-

tal risks that will impair their ability to repay their debts. The

covenant in the Oakland issue is perfectly typical:

The City is unable to predict when seismic events,

fires or other natural events, such as sea rise or other

291. See Richard A. Epstein, Is Global Warming a Public Nuisance?,

OOVER

NST

.: D

EFINING

DEAS

(January 15, 2018), https://www.hoover.org/re

search/global-warming-public-nuisance.

292. See Am. Elec. Power Co. v. Connecticut, 564 U.S. 410 (2011).

293. See Complaint, supra note 288, at ¶ 14.

294. See Epstein, supra note 256.

295. See infra Section III.C.2.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 899

impacts of climate change or flooding from a major

storm, could occur, when they may occur, and, if any

such events occur, whether they will have a material

adverse effect on the business operations or financial

condition of the City or the local economy.

296

It takes no imagination to see the obvious inconsistency

between the two different assertions. But it does raise the hard

question of whether the decision to file the law suits makes the

statements in the bond covenants material misrepresentations

for the purposes of either the Martin Act or the federal securi-

ties law. It is worth noting that the exacting standards of the

Martin Act seem to cover this case even if the truth lies some-

where between the two sets of statements, so that there is some

risk, even if not one to the extent portrayed in these com-

plaints. Given that damages and reliance are not normally re-

quired in Martin Act claims, it looks as though some sanction

should be imposed on these local governments, although it is

by no means clear what it should be: a fine, an upward interest

adjustment, the immediate right to redeem bonds that have

long maturity rates, or something else. It would be instructive

to trace out the prices on these various bonds, but it appears

that they have not moved much in response to the lawsuit.

This suggests that the bond covenants are more accurate than

the allegations in the lawsuits, at which point the other ques-

tion is whether sanctions should be allowed against the plain-

tiffs because of the frivolous nature of their claims. The litiga-

tion is in its earliest stages, but ExxonMobil has already

brought suit in Texas for an order authorizing it “to conduct

depositions and obtain documents pertaining to potential

claims of abuse of process, civil conspiracy, and violations of

ExxonMobil’s constitutional rights.”

297

We can be confident

that this is only the opening salvo in this case, which will go

through many iterations before it reaches a final conclusion,

both on the public nuisance claims and the securities law

claims. But in principle there is no reason why the Martin Act

296. Josh Rosner, Have California Munis Misled Investors and Bond Insurers

About Climate Risk?,

EEKING

LPHA

(Jan. 9, 2018), https://seekingalpha.com

/article/4136231-california-munis-misled-investors-bond-insurers-climate-

risk.

297. Verified Petition for Pre-Suit Depositions of Matthew F. Pawa et al.

(N.D. Tex. 2018) (No. 096-297222-18), https://www.courthousenews.com/

wp-content/uploads/2018/01/ExxonDepositions.pdf.

900 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

should not be a two-edged sword that can be applied with

equal force against parties who understate their potential lia-

bilities under the securities laws. Let us hope that the bond

covenants are entirely accurate, which means that the public

nuisance claims are necessarily incorrect.

Nevertheless, matters have progressed in these various

lawsuits in what will surely be a long and arduous process. In

the New York litigation, the most pointed question from the

bench thus far involved the role of the plaintiffs in bearing any

responsibility for climate change. If the distribution of fossil

fuels with full knowledge of the claimed climate consequences

is improper, then what about the extensive use of the same

fuels by the City of New York in its own activities? Judge Kee-

nan noted that the New York City police, fire and sanitation

departments all have cars and trucks. As Judge Keenan asked

plaintiff’s attorney Matthew Pawa, “Aren’t the plaintiffs using

the product that’s the subject of the lawsuit?” To which Pawa

replied yes.

298

At this point the plaintiffs were in pari delicto—

of equal blame—which could block the claim altogether

under tort law principles. Given that concession, the defen-

dant’s lawyer, Theodore Boutrous, stressed that this problem

did not need a piecemeal, but a global solution.

The litigation in California took a very different tack, as

Judge William Alsup—who has an engineering back-

ground

299

—posted a list of eight questions that spoke to the

full range of historical and technical issues raised by these

cases.

300

These questions were followed by an extended de-

298. Saqib Rahim, NYC tests climate ‘nuisance’ theory in federal court,

OVER-

NOR

’

IND

& S

OLAR

NERGY

OAL

(June 15, 2018), http://governorswind

energycoalition.org/nyc-tests-climate-nuisance-theory-in-federal-court/.

299. Judge Alsup earned a Bachelor of Science in mathematics from Mis-

sissippi State University in 1967 prior to obtaining a J.D. from Harvard Law

School. Alsup, William Haskell,

. J

. C

, www.fjc.gov/history/judges/

alsup-william-haskell (last visited Aug. 16, 2018).

300. The questions are as follows:

1. What caused the various ice ages (including the “little ice age”

and prolonged cool periods) and what caused the ice to melt?

When they melted, by how much did sea level rise?

2. What is the molecular difference by which CO

absorbs infra-

red radiation but oxygen and nitrogen do not?

3. What is the mechanism by which infrared radiation trapped by

in the atmosphere is turned into heat and finds its way

back to sea level?

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 901

bate on March 21, 2018.

301

The plaintiff’s expert witness, Pro-

fessor Myles Allen, got bogged down trying to explain the basic

relationship between increases in CO

and temperature

change under a logarithmic relationship known as Arrhenius’s

law. The defense did not use any expert witness, but followed

the far cleverer strategy of having Chevron’s lawyer Theodore

Boutrous lay out his case with an effective dramatic twist. He

began with the notion that Chevron accepts the consensus of

the scientific community on climate change as announced in

the IPCC,

302

and quickly concluded that “climate change is a

global issue that requires global engagement and global ac-

tion.”

303

Thereafter he teased out many of the uncertainties in

the various iterations of the IPCC’s triennial reports.

304

mentioned that the IPCC noted “economic and population

growth” as drivers of these changes, which result also from

“the cyclical patterns of climate changes over hundreds of

thousands of years” and predated the current increase in CO

4. Does CO

in the atmosphere reflect any sunlight back into

space such that the reflected sunlight never penetrates the at-

mosphere in the first place?

5. Apart from CO

, what happens to the collective heat from tail

pipe exhausts, engine radiators, and all other heat from com-

bustion of fossil fuels? How, if at all, does this collective heat

contribute to warming of the atmosphere?

6. In grade school, many of us were taught that humans exhale

but plants absorb CO

and return oxygen to the air (keep-

ing the carbon for fiber). Is this still valid? If so, why hasn’t

plant life turned the higher levels of CO

back into oxygen?

Given the increase in human population on Earth (four bil-

lion), is human respiration a contributing factor to the buildup

of CO

7. What are the main sources of CO

that account for the incre-

mental buildup of CO

in the atmosphere?

8. What are the main sources of heat that account for the incre-

mental rise in temperature on Earth?

Some Questions for the Tutorial, California v. BP PLC, No. C 17-06011 WHA

(N.D. Cal. Mar. 6, 2018), http://blogs2.law.columbia.edu/climate-change-

litigation/wp-content/uploads/sites/16/case-documents/2018/20180306_

docket-317-cv-06011_order.pdf.

301. See Transcript of Proceedings, California v. BP PLC, No. C 17-06011

WHA (Mar. 21, 2018), https://www.documentcloud.org/documents/4435

400-SF-Oakland-Transcript-of-Alsup-Tutorial-032118.html.

302. Id. at 80.

303. Id. at 82.

304. Id. at 90–91.

902 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

attributable to human beings starting in about 1950.

305

It was

these changes that contributed to the Little Ice Age, which the

IPCC report attributes to such issues as lower solar activity and

volcanic activities—neither of which involves changes in CO

concentrations.

306

In so doing he sketched out an argument

that separated climate change from changes in carbon dioxide

levels.

On June 25, 2018, Judge Alsup dismissed the California

and Oakland litigation in a pointed sixteen-page opinion for

failure to state a claim.

307

After a detailed review of the sci-

ence presented at the March 21 proceedings, including find-

ing Arrhenius was concerned with global cooling rather than

warming,

308

the opinion treated the underlying science of

global warming as a given: “This order fully accepts the vast

scientific consensus that the combustion of fossil fuels has ma-

terially increased atmospheric carbon dioxide levels[.]”

309

Elsewhere, he acknowledged “other causes” were also at

work.

310

Nevertheless, perhaps reflecting the wisdom of Chev-

ron’s litigation strategy, the issue in the case “is not over sci-

ence” but was a “legal one.”

311

Applying the basic tort law of

public nuisance, Judge Alsup found the question came down

to balancing the historical civilizational utility of extracting fos-

sil fuels against the real harm incurred by plaintiffs to deter-

mine whether the defendants’ conduct was reasonable.

312

His

order held that the grand global nature of this policy question

makes it a federal one, but because of its gargantuan eco-

nomic and foreign relations implications, one for the political

branches to determine, not the courts.

313

Judge Alsup’s order

thus refrained from weighing in on the substantive scientific

305. Id. at 94.

306. Id. at 98.

307. Order Granting Motion to Dismiss Amended Complaints, City of

Oakland v. BP PLC, No. C 17-06011 WHA (N.D. Cal. June 25, 2018), http://

blogs2.law.columbia.edu/climate-change-litigation/wp-content/uploads/

sites/16/case-documents/2018/20180625_docket-317-cv-06011_order-1.pdf.

308. Id. at 3.

309. Id. at 12.

310. Id. at 4.

311. Id. at 6.

312. Id. at 7–8 (citing

ESTATEMENT

ECOND

)

ORTS

§§ 826–832 (

NST

1979)).

313. Id. at 14–16.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 903

issues discussed above. But it does suggest a blow to activists

who continue to press suits in New York and elsewhere.

It is still unclear whether these remaining cases will be set

for trial or dismissed, perhaps on the grounds that the need

for a comprehensive solution is inconsistent with the piece-

meal relief that is requested. But for these purposes, what is

most apparent is that the common law, the Martin Act, and

the federal securities acts all give ample room to raise these

issues—and to reject the claims once they are raised. What

matters ultimately are the scientific issues about which Bout-

rous discussed in his exchange with Judge Alsup. It is to those

that I turn next.

2. The Science

It is clear that the various actions against ExxonMobil

under the Martin Act and against multiple oil companies

under the law of public nuisance raise a common core of sci-

ence issues that require some independent analysis. The

Schneiderman position in March 2016 echoed a common

theme in this literature, which is that the science is so settled

that the only question that remains is to devise a sensible solu-

tion to the problem. In all these cases, it seems as though indi-

vidual litigation is a poor way in which to proceed. But

whether or not this matter is returned to the legislative arena,

as suggested by Judge Alsup’s opinion, it is important to ask

whether the science is as solid in one direction as is commonly

supposed.

The answer is that it is not. Right now, many climate

scientists take issue with the question of whether global warm-

ing attributable to man-made CO

poses a substantial threat,

backed by extensive evidence. Patrick Michaels and Paul Knap-

penberger articulate a view, hinted at by the title of their re-

cent book, Lukewarming, that any increase in CO

levels could

be expected, with other things held equal, to produce some

modest increase in global temperatures which do not a pre-

sent a significant environmental danger.

314

The CO

Coali-

tion, which I have worked with from time to time, goes one

step further and claims that the recent increases in CO

levels

314. See

ATRICK

J. M

ICHAELS

& P

AUL

C. K

NAPPENBERGER

UKEWARMING

LIMATE

CIENCE

HAT

HANGES

VERYTHING

116–17 (2016).

904 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

result in a net benefit to humanity.

315

From literally hundreds

of articles that I have reviewed on both sides of the debate, it

becomes all too clear that there are few people taking a

maybe-yes/maybe-no stance. On virtually every point, the

global warming debate gives rise to fierce hand-to-hand com-

bat that cannot easily be put to rest, and certainly not so by the

collection of factoids that Schneiderman and Pawa strung to-

gether in rapid fashion. It is, therefore, worthwhile to step

back from his charges of fraud to ask whether the evidence on

the scope and extent of global warming is as unequivocal as he

claims. In the context of the Martin Act, it is not necessary for

these purposes to show that the many reputable scientists are

correct in claiming that fears of global warming are over-

wrought. It is simply sufficient to show that there are two sides

to the argument.

The first point is that one has to be careful of the com-

mon claim that 97% of scientists believe in global warming.

There are good reasons to believe that this figure misstates the

division of authority within the scientific community.

316

see why this is the case, it is critical to isolate the key ambiguity

inherent in this formulation. As commonly stated, the 97%

consensus includes in its ranks any scientist who thinks that

some increase in CO

concentrations will lead to some increase

in temperature, a broad and undefined class which includes

most people who are self-described climate skeptics. Indeed,

this particular claim appears to be a necessary truth. But the

harder questions ask how much of an increase in CO

poses a

315. See

OALITION

, http://co2coalition.org (last visited Aug. 12,

2018). The organization has been attacked as a pawn of the fossil fuel indus-

try. See, e.g., CO

Coalition,

ESMOG

, http://www.desmogblog.com/co2-coali

tion (last visited Oct. 29, 2016); John Cook et al., Quantifying the Consensus on

Anthropogenic Global Warming in the Scientific Literature, 8

NVIRON

. R

. L

ET-

TERS

1 (2013), http://iopscience.iop.org/article/10.1088/1748-9326/8/2/

024024/pdf (“Our analysis indicates that the number of papers rejecting the

consensus on AGW is a vanishingly small proportion of the published re-

search.”). The study was confined to an analysis of abstracts, without any

effort to look into the merits of any of the 11,944 papers reviewed.

316. See Joseph Bast & Roy Spencer, The Myth of the Climate Change ‘97%’,

ALL

. J.

(May 26, 2014, 7:13 PM), https://www.wsj.com/articles/joseph-

bast-and-roy-spencer-the-myth-of-the-climate-change-97-1401145980; see also

Judith Curry, The 97% ‘consensus’,

LIMATE

(July 26, 2013)

https://

judithcurry.com/2013/07/26/the-97-consensus/.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 905

threat and why. That point has been repeatedly hammered

home by others, including Matt Ridley, who has written:

The supposed 97% consensus, based on a hilariously

bogus study by John Cook, refers only to the proposi-

tion that climate change is real and partly man-made.

Nobody has ever shown anything like a consensus

among scientists for the proposition that climate

change is going to be dangerous.

Professor Daniel Sarewitz put it well recently: “Even

the vaunted scientific consensus around climate

change . . . applies only to a narrow claim about the

discernible human impact on global warming. The

minute you get into questions about the rate and se-

verity of future impacts, or the costs of and best path-

ways for addressing them, no semblance of consensus

among experts remains.”

317

These observations are not isolated. Recent surveys about

the attitude of members of the American Meteorological Soci-

ety “found that, while a majority of meteorologists surveyed are

convinced humans have contributed to global warming [64%],

this was a substantially smaller majority than that found among

all Earth scientists (82%).”

318

It also found that “[s]ome mem-

bers have expressed that their views, which question the view

that human-caused global warming was occurring, are treated

with hostility within the [American Meteorological Soci-

ety].”

319

It is pointless to dwell on political animosity, but it is criti-

cal to properly frame the conceptual issue discussed before

317. Matt Ridley, Global Warming Versus Global Greening,

LOB

. W

ARMING

’

OUND

(Oct. 18, 2016), http://www.thegwpf.org/matt-ridley-global-

warming-versus-global-greening/ (citing Cook et al., supra note 315). See also

Daniel Sarewitz, Saving Science,

TLANTIS

(2016),

http://

www.thenewatlantis.com/publications/saving-science. Sarewitz rightly relies

on Vannevar Bush’s 1945 report to President Truman, which charted the

course for post-war advances in scientific research. See

ENNEVAR

USH

CI-

ENCE

NDLESS

RONTIER

: A R

EPORT TO THE

RESIDENT

(1945), https://

www.nsf.gov/od/lpa/nsf50/vbush1945.htm.

318. Neil Stenhouse et al., Meteorologists’ Views About Global Warming: A Sur-

vey of American Meteorological Society Professional Members, 95

ULL

. A

. M

ETEOR-

OLOGICAL

’

1029, 1029

(2014). Other surveys found a high fraction of

scientists, often around 80 percent, believe that “induced climate change is

occurring,” without, apparently, determining its likely severity. Id.

319. Id. at 1030.

906 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

Judge Alsup. To do so, I rely on the clear exposition of the

problem by the CO

Coalition. The standard formula for

global warming rests on two key variables. The first of these is

the rate of expected increase in CO

concentrations. The sec-

ond is the size of a sensitivity coefficient, or “S.” As to the first,

the impact of temperature changes varies with the logarithmic

change in concentration, which always rises more slowly than

the percentage increase, such that a doubling of concentra-

tion leads to about a 31% increase in temperature. But an in-

crease of what? On this last point, the sensitivity index mea-

sures the rate at which the higher concentrations translate into

higher temperatures. The higher the level of sensitivity, the

greater the response will be to any given increase (or de-

crease) in the level of concentration of CO

. The setup, often

called the Arrhenius equation, is as follows:

T = S log

(C2/C1), where T = T2 – T1.

The proposition that higher concentrations of CO

leads

to some degree of global warming only requires that S, the

doubling sensitivity number, be greater than zero for any in-

crease in CO

concentration to increase global temperature.

So long as any of that CO

is produced by human beings, the

proposition that human action causes global warming is (trivi-

ally) true, but entirely consistent with the Lukewarming hypoth-

esis and, indeed, the possibility that CO

produces net social

benefits.

The key question, therefore, for the purposes of the

global-warming debate, is just how much greater than zero is

S. Where S is low, the temperature increases are small. Where

S is large, the temperature increases are obviously greater. In

all cases, the increases get smaller in percentage terms as the

concentrations gets higher, given the logarithmic nature of

the basic relationship. The key scientific dispute in this regard

is about the magnitude of S. The conventional models, includ-

ing those used by the Intergovernmental Panel on Climate

Change (IPCC), a leading international, UN-sponsored body

dedicated to measuring the effects of climate change, tend to

place S > 3 or so, which is a high number. Others think that

more recent evidence puts S far lower, perhaps at S < 2, at

which point the increases in temperature are far slower.

320

320. See Robert P. Murphy, Patrick J. Michaels & Paul C. Knappenberger,

The Case Against the Carbon Tax (CATO Inst., Working Paper No. 33, 2015),

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 907

This factor has in turn been challenged as being too low.

321

One reason to incline toward the lower estimate is that water

vapor, also a greenhouse gas, has at least some influence on

temperature, which necessarily means that CO

should matter

less proportionally. There is, moreover, no obvious monotonic

relationship between water vapor and temperature, which

means that shifts in its concentration over time and over differ-

ent portions of the earth can explain some of the variability in

regional responses that models based on CO

levels cannot,

i.e., periods, such as that in the middle of the 1970s, where

cooling rather than warming has been observed.

On this score, it is instructive to look at the graph that

Professor Richard Lindzen offers of temperatures since 1920:

FIGURE 1

CHEMATIC OF

EMPERATURE

NOMALY

ECORD

Amazingly, 18 of the 18 warmest years on record occurred during last 18 years

http://object.cato.org/sites/cato.org/files/pubs/pdf/cato-working-paper-

33.pdf.

321. See Joseph Majkut, Is There a Divergence Between Climate Models and Tem-

perature Data?,

ISKANEN

.: C

LIMATE

& E

NERGY

’

(Dec. 1, 2015), http:

//niskanencenter.org/blog/is-there-a-divergence-between-climate-models-

and-temperature-data/.

908 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

During the same period, CO

levels moved steadily up-

ward from about 320 ppm in 1960 to somewhat over 400 ppm

today. As Lindzen has observed, “the warming episode from

about 1978 to 1998 appeared to have ceased and temperatures

have remained almost constant since 1998.”

322

As he then

notes, it follows necessarily that all of the temperature reports

in last 18 years have been close to the high point, given that

there has been no decline in temperature. The monotonic in-

creases in the level of CO

, whether natural or man-made, can-

not explain why temperatures first rose, then declined, and

then rose again, only to remain flat thereafter. Any single

driver for temperature thus has to be ruled out on this data.

levels surely play some role, but so must other factors.

The approximately 13 percent increase in CO

levels over the

last 18 years has generated at most only modest atmospheric

temperature increases, which raises the claim, strongly denied

by others, of a “pause” in global warming.

323

Indeed, there is

scant evidence that even major reductions in the overall level

of CO

are likely to have an appreciable effect on long-term

global warming in light of the interplay of social forces.

324

Per-

haps some insidious effects of CO

may be delayed, but at the

very least those additional complications cast doubt on

Schneiderman’s categorical conclusions.

322. Richard Lindzen, Thoughts on the Public Discourse over Climate Change,

ERION

EST

(

Apr. 25, 2017), http://merionwest.com/2017/04/25/rich

ard-lindzen-thoughts-on-the-public-discourse-over-climate-change/.

323. For rival views, see Brian Kahn, No Pause in Global Warming,

. A

(June 4, 2015), http://www.scientificamerican.com/article/no-pause-in-

global-warming/ (“The global warming hiatus—a decade-plus slowdown in

warming—could be chalked up to some buoys, a few extra years of data and

a couple buckets of seawater.”); John C. Fyfe et al., Making Sense of the Early-

2000s Warming Slowdown, 6

ATURE

LIMATE

HANGE

224, 224 (2016) (“It

has been claimed that the early-2000s global warming slowdown or hiatus,

characterized by a reduced rate of global surface warming, has been over-

stated, lacks sound scientific basis, or is unsupported by observations. The

evidence presented here contradicts these claims.”).

324. The U.S. contribution to reduced global temperatures in 2100 would

be about 0.03 of a degree, using the EPA’s own climate model, under as-

sumptions that exaggerate the effectiveness of the policies. Note that the

standard deviation of the temperature record is about 0.1 of a degree, so

that the U.S. effect would not be measurable against normal variation. See

Benjamin Zycher, The Carbon Tax is Not Just Political; It’s Ineffective, Too,

ILL

(Sept. 28, 2016, 2:01 PM), http://thehill.com/blogs/pundits-blog/en

ergy-environment/298285-the-carbon-tax-is-not-just-political-its-ineffective.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 909

The rise and fall of global temperatures is also confirmed

by a comparison of the general patterns of increase for the

years between 1895 and 1946 and between 1957 and 2008.

325

To be sure, the temperature range in the earlier period was

slightly lower, but the up-and-down patterns of temperature

changes were the same in the low-CO

environment for the

early period and high-CO

environment for the latter one,

again showing that multiple factors are necessarily in play. In

this complex environment, it is not credible that a single

driver could have the high level of influence attributable to

. One countervailing force that tends to reduce tempera-

ture increase is the release of aerosols, whose presence in the

atmosphere tends to counteract any influence of CO

326

long as those move independently of changes in CO

concen-

trations, their combined effect is difficult to determine no

matter how long or short the time horizon. NYAG Schneider-

man also pointed to the shrinking of Arctic Sea ice mass, an

assertion commonly made in authoritative places.

327

But re-

cent accounts point strongly in the opposite direction. One

recent report recounts “[s]ince hitting its earliest minimum

extent since 1997, Arctic sea ice has been expanding at a phe-

nomenal rate,”

328

noting that Arctic Sea ice levels have in-

creased by about 25% between 2012 and 2016.

329

Historically,

moreover, Arctic Sea ice has sometimes been at much lower

325. See

OALITION

, A P

RIMER ON

ARBON

IOXIDE

6 (2016), http://

co2coalition.org/wp-content/uploads/2016/08/Primer-on-CO2-and-Cli

mate-1.pdf.

326. Id. at 7.

327. See, e.g., Arctic Sea Ice Minimum

, NASA: F

ACTS

, https://climate.nasa

.gov/vital-signs/arctic-sea-ice/ (last visited Aug. 12, 2018).

328. Paul Homewood, Record Arctic Sea Ice Growth in September,

OT A

EOPLE

NOW

HAT

(Sept. 25, 2016), https://notalotofpeopleknowthat.

wordpress.com/2016/09/25/record-arctic-sea-ice-growth-in-september/.

The charts are always capable of multiple interpretations, but one useful

point is that the extent of the Arctic Sea ice today is about 33% higher than

it was at the same date in 2012, which in turn was only about 75% of what it

was at its 2001 peak. The movement over the last ten years has been roughly

stable. See also Thomas Richard, Arctic Sea Ice Rebounds as Polar Temperatures

Start to Plunge,

LASTING

EWS

(Sept. 19, 2016), http://us.blastingnews.com/

news/2016/09/arctic-sea-ice-rebounds-as-polar-temperatures-start-to-plunge-

001129573.html?utm_source=CFACT+Updates&utm_campaign=4c96e5bd2

3-Call_off_the_death_spiral_9_19_2016&utm_medium=email&utm_term=0_

a28eaedb56-4c96e5bd23-260145993.

329. Homewood, supra note 328.

910 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

levels, so that in the 1903–1908 period, famed Arctic explorer

Roald Amundsen was able to navigate a northwest passage

through the Arctic by boat.

330

The cyclical nature of Arctic Sea

ice seems to be largely independent of CO

concentrations, so

that the recent record increases in CO

concentrations have

seen both ebbs and flows in Arctic Sea ice concentration.

Indeed, most of the alarmist predictions have been off

base. Oftentimes this position is denied categorically, as in the

open letter prepared by MIT President, L. Rafael Reif, pro-

testing the decision of Donald Trump to withdraw from the

Paris Accords on global warming. Reif led with this categorical

statement: “As human activity emits more greenhouse gases

into the atmosphere, the global average surface temperature

will continue to rise, driving rising sea levels and extreme

weather.”

331

Reif’s letter offered no evidence for this proposi-

tion and it prompted the following pointed response on these

points from Willie Soon and Christopher Monckton:

The average sea level rise since 1870 has been 1.3–1.5

mm (about a twentieth of an inch) per year, or five

inches per century. Professor Nils-Axel M

orner, a re-

nowned sea-level researcher who has published more

than 500 peer-reviewed articles on this topic, has

been unable to find observational evidence that sup-

ports the models’ predictions of dramatically acceler-

ating sea level rise.

Observations over the last few decades indicate that

extreme weather events, including tornadoes and

hurricanes, have been decreasing, rather than increas-

ing, both in number and in intensity. Moreover, total

accumulated cyclonic energy has also been declining.

As MIT Emeritus Professor Richard Lindzen has ex-

plained, the decline in storminess is a consequence

of reduced temperature differentials between the

tropics and exo-tropics that arise when global average

temperatures are slightly warmer.

330. Roald Amundsen, 15

NCYCLOPEDIA

RCTICA

: B

IOGRAPHIES

18–24

(1951), http://collections.dartmouth.edu/arctica-beta/html/EA15-05.html.

Amundsen made the trip in a 46-ton fishing boat. The trip took three years,

but clearly the Arctic ice had to melt during it. Id. at 19, 20–22.

331. Letter from L. Rafael Reif, President, Mass. Inst. Tech., to MIT Com-

munity (June 2, 2017), http://news.mit.edu/2017/letter-mit-community-us-

withdrawal-paris-climate-agreement-0602.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 911

Looking at the United States, major hurricane activ-

ity is at a record low. As of June 1, 2017, it had been

eleven years and seven months since a category 3 to 5

hurricane last struck the U.S. mainland. According to

NOAA Hurricane Research Division data, the previ-

ous record was nine years, set in 1860–1869.

332

Those general observations on sea level are borne out by

comparing predictions of sea level change with the observed

data since 1996—a period which has seen substantial increases

in carbon dioxide levels.

333

IGURE

LOBAL

EVEL

332. Willie Soon & Christopher Monckton of Brenchley, Response to MIT

President: Paris Exit Scientifically Sound (Part II),

ASTER

(July 6, 2017),

http://www.masterresource.org/climate-science/mit-president-exit-paris-ii/.

I supported the President’s decision on scientific, not political grounds. See

Richard A. Epstein, Forget the Paris Accords,

OOVER

NST

: D

EFINING

DEAS

(May 30, 2017), http://www.hoover.org/research/forget-paris-accords;

Richard A. Epstein, Containing Climate Change Hysteria,

OOVER

NST

.: D

EFIN-

ING

DEAS

(June 5, 2017), http://www.hoover.org/research/containing-cli

mate-change-hysteria.

333. This figure, and others of similar import are available through a sim-

ple Yahoo Images search for the term “ocean level rise chart.”

912 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

Other studies report similar results, including a rejection

of the claim that global warming increases general ecological

instability, which in turn increases the number of extreme

events, like hurricanes and cyclones. Instead the data tend to

show that these events have been in a modest, but consistent,

decline over the last century.

334

Indeed, between Hurricane

Wilma in 2005 (just a few weeks after the far more devastating

Hurricane Katrina) and Hurricane Harvey in 2017, the United

States experienced a decade of relatively modest hurricane ac-

tivity resulting in one of the quietest periods since the data has

been collected,

335

shattered only by the devastating down-

pours from Hurricane Harvey in late August 2017.

336

But once

again it would be a mistake to attribute this behavior to in-

crease in CO

levels, for it is well established that the cycles

between El Ni

no (which sends cold air west to the Atlantic)

and La Ni

na (which does not) have much to do with the move-

ment of hurricanes from the West Coast of Africa to the Carib-

bean and the Southern United States.

337

That phenomenon is

driven by the so-called El Ni

no Southern Oscillation (ENSO),

which involves the constant cycling of surface ocean tempera-

tures in the Pacific equatorial region that drives climate pat-

terns across both oceans.

The skeptical account of the role of CO

in global warm-

ing receives additional support from one finding that embar-

rasses the global-warming hypothesis. A compilation of about

100 estimates of global warming, when applied over the period

334. See

OALITION

, supra note 325, at

17–

19. Despite the late twenti-

eth century warming period, the data show no upward trend in strong torna-

does; in fact, there appears to be a slight downward trend in the frequency

of strong tornadoes. See id. at 19.

335. Without discounting the effects of other storms such as Superstorm

Sandy in 2012 (which made landfall as a Category 2 storm), there were no

Category 3 or above hurricanes to make landfall in the United States for

nearly a decade until Hurricane Harvey. See Complete List of Continental U.S.

Landfalling Hurricanes,

’

CEANIC

& A

TMOSPHERIC

DMIN

.: H

URRICANE

. D

(June 1, 2017), http://www.aoml.noaa.gov/hrd/tcfaq/E23.html.

336. Erin Ailworth, Dan Frosch & Christopher M. Matthews, Harvey to Hit

Flooded Houston Again,

ALL

. J

. (Aug. 28, 2017, 7:07 AM), https://

www.wsj.com/articles/harvey-to-hit-houston-again-1503927231.

337. For a brief account, see Aylin Woodward, La Ni

na Forecast May Mean

Even Worse Atlantic Hurricanes in 2018,

CIENTIST

(Oct. 25, 2017), https:/

/www.newscientist.com/article/2151474-la-nina-forecast-may-mean-even-

worse-atlantic-hurricanes-in-2018/.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 913

between 1979 and 2013, predicted an increase of 1

C, when

the actual result for this period is about 0.2

C to 0.25

C—a

four- to five-fold disparity, depending on whether satellite or

balloon measurements are used.

338

Surely a deviation between

prediction and reality of that magnitude cries out for fresh

thinking about the underlying problem. Since measures of

increase can be accurately observed, the key inquiry must

surround the proper determination of S, which is not directly

observable. But a better fit for the data comes from noting that

S is on the lower side of the relevant estimates. Nothing that

Schneiderman and other defenders of global warming have

said addresses these difficulties. It cannot simply be assumed

that the IPCC has made the best estimate of S.

Even if recent temperature increases are modest at best,

the same is not true of the level of greenery on the earth’s

surface. Matt Ridley, in his lecture “Global Warming Versus

Global Greening,”

339

offers powerful evidence that the main

consequence of CO

has been an increased level of global

greening, to the tune of 14% of land coverage over the last

thirty years. That one number, if true, suggests that the so-

called social cost of carbon may well be, at least for restricted

ranges, negative and not positive.

340

A group of Australian

scientists also published an earlier study that predicted

“[u]sing gas exchange theory” that a “14% increase in atmos-

pheric CO

(1982–2010) led to a 5 to 10% increase in green

foliage cover in warm, arid environments. Satellite observa-

tions, analyzed to remove the effect of variations in precipita-

tion, show that cover across these environments has increased

by 11%.”

341

In this instance, there is ample reason to believe

that these large greening effects are attributable to increases

in CO

, given that in controlled experiments, plant growers

338. For the graph, see Climate Science: Assumptions, Policy Implications, and

the Scientific Method: Hearing Before the H. Comm. of Sci., Space & Tech., 115th

Cong. 35–51 (2017) (Testimony of John R. Christy).

339. Matt Ridley, Lecture at the 2016 Global Warming Policy Forum:

Global Warming versus Global Greening (Oct. 17, 2016), http://www.the

gwpf.com/matt-ridley-global-warming-versus-global-greening/.

340. Id.

341. Randall Donohue et al., Impact of CO

Fertilization on Maximum Foliage

Cover Across the Globe’s Warm, Arid Environments, 40

EOPHYSICAL

. L

ETTERS

3031 (2013), https://openresearch-repository.anu.edu.au/bitstream /1885/

73415/2/01_Donohue_Impact_of_CO_2__fertilization_2013.pdf.

914 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

pump in CO

to increase growth rates. Other correlated vari-

ables include temperature, precipitation and radiation.

342

Studies have also found that rising CO

levels can lead plants

to use water more efficiently, increasing vegetation every-

where, in particular the dry areas.

343

To let CO

in, plants have to open up pores called

stomata in their leaves, which in turn allows water to

sneak out. Plants thus need to strike a balance be-

tween taking up carbon to build new leaves, stems

and roots, while minimizing water loss in the process.

This led to sophisticated adaptations that ha[ve] al-

lowed many plant species to conquer a range of arid

environments.

344

To make this analysis even more comprehensive, it is criti-

cal to take into account the interplay of both global and local

forces in any particular case. Thus, a claim that water levels are

rising in Florida could be attributable to water rising or to land

sinking, or some combination of both. Nor is it clear how

much of the sea level increase can be attributed to global

warming, when the rise has been relatively constant at seven

inches per century for the last 3000 years.

345

Similarly, the

large calving of an iceberg from the Antarctica cannot be at-

tributable to tiny changes in global temperatures, especially

342. Zaichun Zhu et al., Greening of the Earth and Its Drivers,

NATURE

, http:/

/www.nature.com/articles/nclimate3004.pdf. But see Carbon Dioxide Fertiliza-

tion Greening Earth, Study Finds, NASA, https://www.nasa.gov/feature/god

dard/2016/carbon-dioxide-fertilization-greening-earth (noting that rising

carbon dioxide concentrations is the chief culprit of global warming and its

beneficial impact on plants may be limited).

343. Rising Carbon Dioxide is Making the World’s Plants More Water-Wise,

ONVERSATION

, http://theconversation.com/rising-carbon-dioxide-is-mak

ing-the-worlds-plants-more-water-wise-79427 (last visited Apr. 13, 2018).

344. Id. (also pointing out that the water savings could lead to increased

water runoff, and precipitation may not go through vegetation but through

direct soil evaporation, which might not necessarily benefit humans in the

end).

345. Larry Bell, Is Climate Change Elevating Coastal Flooding Risk?,

CFACT

(Sept. 26, 2016), https://www.cfact.org/2016/09/26/is-climate-change-ele

vating-coastal-flooding-risk/?utm_source=CFACT+Updates&utm_campaign=

49124bbde3-Flood_of_disinformation9_28_2016&utm_medium=email&utm

_term=0_a28eaedb56-49124bbde3-260145993. On this point, the Union of

Concerned Scientists appears to overstate the case for attributing rising sea

levels to global warming. See An Open Letter Regarding Climate Change from Con-

cerned Members of the U.S. National Academy of Sciences, supra note 250.

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 915

when the size of the ice pack is increasing elsewhere on Ant-

arctica. Rapid movements in ice are much more likely to be

the result of localized volcanic activity beneath the surface.

In addition, many local losses can be mitigated by undo-

ing key man-made adaptations. The Sierra Club has made just

this point when it writes, “[w]hat we can and must do is to let

nature resume control of water flow by removing barriers like

the Tamiami Trail, the Miami Canal and the L-67 canals, while

cleaning up Big Sugar’s effluent to Everglades standards.”

346

Much of the runoff in the Everglades comes from the con-

struction of a network of oil exploration channels before

World War II. The rapid flow of water through these channels

disrupts the natural ecology. Filling them up should not be a

huge problem, but it has never been undertaken. In addition,

the runoff of phosphorus from nearby farms undermines the

local environment.

347

There is also an extensive empirical dispute over both

long and short-term trends in global warming, wholly apart

from the explanation of its sources. On this score, recent re-

search suggests that we are now at the end of a global warming

period extending close to 11,000 years—the Holocene epoch,

during which time temperatures fell gradually as CO

levels

increased, a finding that is in obvious tension with Schneider-

man’s claims in his March 2016 press conference. Once again

there must be some natural force to explain a relationship that

is the inverse of what is normally expected.

348

The more con-

troversial portion of the long-term research swirls around a fa-

mous doomsday proposal in 2000 from Michael Mann and his

coauthors. These authors posit a “hockey stick” version of

human history that sees very stable temperatures over hun-

dreds of years, followed by a sharp increase in temperatures

around 1900, roughly corresponding to the major uptick in

346. Jon Ullman, The Everglades and Sea Level Rise: The Clock Is Ticking,

I-

ERRA

LUB

. N

EWS

(Feb. 26, 2013), http://www.sierraclubflorida

news.org/2013/02/the-everglades-and-sea-level-rise-clock.html.

347. Phosphorus: Challenges Part 3,

RIENDS OF THE

VERGLADES

https://

www.everglades.org/phosphorous (last visited Aug. 16, 2018) (“Phosphorus

from agricultural runoff has impaired water quality in large portions of the

Everglades and has been particularly problematic in Lake Okeechobee.”).

348. See Andy May, Lindzen, Soon and Spence Debunked?,

ATTS

ITH

HAT

(May 3, 2017), https://wattsupwiththat.com/2017/05/03/lindzen-

soon-and-spencer-debunked/.

916 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

the use of fossil fuels.

349

It is easy to extrapolate from that

model ruinous temperature increases in the next century that

could fully justify the most alarmist predictions possible.

But there is, however, the major question as to whether

the model is true. On this score, Mann seems to think the best

way to firm up his position is by bringing defamation suits

against those who disagree. One such suit is against the writer

Mark Steyn who claimed that Mann “has perverted the norms

of science on an industrial scale”—a slight that Mann would

have been well-advised to ignore.

350

More instructive is Mann’s

litigation in Canada against the Canadian climatologist Tim

Ball claiming “Mann belongs in the state pen, not Penn.

State,” for fraudulent misstatement of data.

351

This case took

an odd twist when Mann claimed that he did not have to re-

lease his raw data because it was his intellectual property—a

claim that is clearly wrong for at least two reasons. First, the

property comes from a government project for which data

generally has to be made public wholly apart from litigation.

Second, turning over the documents and data to the court

does not imply that the recipient may publish them because

they are first subject to judicial inspection to see if a more se-

lective claim of privilege can be sustained, and then, if need

be, turned over to the plaintiffs under some kind of protective

order.

352

Scientifically, however, the most striking point is the com-

parison of the historical account offered by Mann—which has

low variability until the onset of the 20th century—and the

Ball version, which has far higher variability in earlier periods

and relatively modest level of temperature variability (wholly

apart from any causation question) for the post-1800 period,

349. Michael E. Mann, Raymond S. Bradley & Malcolm K. Hughes, Global-

Scale Temperature Patterns and Climate Forcing Over the Past Six Centuries, 392

ATURE

779 (

Apr.

23, 1998

). For a vigorous modern defense of Mann, see

Chris Mooney, The Hockey Stick: The Most Controversial Chart in Science, Ex-

plained,

TLANTIC

(May 13, 2013), http://www.theatlantic.com/technol

ogy/archive/2013/05/the-hockey-stick-the-most-controversial-chart-in-scien

ce-explained/275753/.

350. John O’Sullivan, Breaking: Fatal Courtroom Act Ruins Michael ‘Hockey

Stick’ Mann,

RINCIPIA

. I

’

(

July 4, 2017), http://principia-scientific.

org/breaking-fatal-courtroom-act-ruins-michael-hockey-stick-mann/.

351. Id.

352. See

. R. C

. P.

26(b)(2)(B).

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 917

smaller than that for the 900 to 1200 A.D. period.

353

See for

yourself:

IGURE

ATTLE OF THE GRAPHS

: M

ANN VERSUS

ALL

Obviously, if Ball’s graph is accurate, then Mann’s is seri-

ously misleading. There is evidence of a major amount of

warming during the medieval period, which is best captured

by the accurate historical evidence that both Greenland and

Labrador had temperate climates in which grapes could grow.

That fits in neatly with the Ball model, and in fact was previ-

ously adopted in the IPCC’s 1990 Report.

354

But it is much

more difficult to explain under the Mann model. It is of

course possible that the temperature changes in the North At-

lantic were not representative of overall global changes. This

point is surely debatable. Nevertheless, there are at least two

macro-level difficulties with the Mann model. First, the only

way to preserve the model’s low level of variability to postulate

a simultaneous large decrease in temperatures elsewhere that

353. See Mann, Bradley & Hughes, supra note 349.

354. For the various maps, one can do a simple Google image search to

see the variability. See

OOGLE

MAGES

, https://www.google.com/images

(search for “medieval warm period little ice age graph”).

918 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

parallel the increase in temperatures in the North Atlantic.

The second is that any such model of variability makes it

highly unlikely that changes in CO

levels, whose impacts are

worldwide, could produce both effects at the same time. The

puzzle is even greater because the change in CO

levels during

this period are too small to count as an effective driver. Hence,

the data would require finding other explanations, which in

turn, reduces the overall importance of CO

. All readers

should judge for themselves who has the better of the debate,

a judgment that is always subject to revision. But neither the

Martin Act nor the federal securities laws have anything useful

to contribute to this debate.

ONCLUSION

In dealing with the issues under the Martin Act, it is criti-

cal to deal with both the procedural and substantive issues. In

general, most of the dangers lie in the unique procedural fea-

tures of the Martin Act, not its substantive provisions.

The chief vice of the Martin Act stems from the incredible

discretion that it confers on the NYAG to conduct either pub-

lic or private investigations of potential targets, without limita-

tion. Similarly, broad investigative powers are also available to

the SEC, which give it, like other federal agencies, the power

to initiate lawsuits that can bankrupt major corporations and

require them to forfeit their licenses when indicted for partic-

ular offenses. There are of course important differences in

that the SEC has to act by majority vote of a five-member com-

mission while the NYAG can move unilaterally. But even here,

it is important to recall the SEC has three commissions from

the President’s party and two from the opposition, so that

votes rigidly divided along partisan lines can often lead to in-

vestigations and indictments, which are exceedingly difficult to

resist given that the sanctions for being indicted, including

loss of licenses, are more devastating than any fine that might

be imposed at the end of the ordeal. It was just this unfortu-

nate state of affairs that drove Arthur Andersen over the brink.

But it is critical to note that this was not by the SEC, which has

no authority to bring criminal prosecutions, but by the Depart-

ment of Justice, which could, like the NYAG, mount criminal

prosecution on a set of dubious charges that lacked the critical

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 919

element of scienter.

355

The parallel actions by the SEC are, at

least in theory, cabined by its rulemaking. These actions are

subject to the Administrative Procedure Act,

356

as well as lim-

ited by statutory language that requires the SEC rules to con-

sider “whether the action will promote efficiency, competition,

and capital formation.”

357

Thus the lack of institutional safe-

guards makes civil and criminal prosecutions far easier in cases

that stray far from the financial fraud cases to which the Mar-

tin Act was originally directed. That precise result seemed to

be in formation with the ill-advised efforts of NYAG Eric

Schneiderman to invoke the investigative powers of the Martin

Act to challenge the activities of ExxonMobil over the past

forty years, on evidence that turns out to be fatally weak.

If these procedural shortfalls require prompt correction,

the picture with the substantive requirements is far more

nuanced. The focus of the inquiry was whether the law’s de-

parture from the substantive common-law requirements of sci-

enter, reliance, and damages had brought about a fatal imbal-

ance in the enforcement of the law against major public com-

panies and their officials. In general, I think that critique is

overblown for at least two reasons, and that the greater danger

by far lies in the excessive concentration of power lodged in

the NYAG. Turning first to the substantive issues, the removal

of these limitations did not, in practice, create a free-for-all

that allowed the NYAG to sue any firm or executive for any

reason. Even after the passage of the Martin Act, the addi-

tional requirements used to cabin potential liability for false

words still remained in place, so that, substantively, the Martin

Act on many key points bears a close resemblance to both

common-law fraud and federal securities laws, which work in

the same area. Thus, under the Martin Act, the state is not

spared the burden of proving a false statement of fact, as op-

posed to a statement of opinion or the prediction of a future

event. In addition, it is necessary for the state to prove that the

355. See Arthur Andersen LLP v. United States, 544 U.S. 696 (2005) (over-

turning a charge of obstruction of justice for want of the proof of scienter).

For discussion of the undue leverage of the government in these

prosecutorial situations, see Epstein, Deferred Prosecution Agreements on Trial,

supra note 30.

356. Administrative Procedure Act, 60 Stat. 237 (1946) (codified as

amended at 5 U.S.C. §§ 701–708 (2012)).

357. 15 U.S.C. § 78c(f).

920 NYU JOURNAL OF LAW & BUSINESS [Vol. 14:805

false statements were material, so that they were likely to exert

a real influence on the behavior of the parties to whom they

were directed. It was also incumbent to show that the state-

ments were made incident to the purchase or sale of a security,

or of an offering to buy or sell securities, which precluded its

application to general public statements that might in some

indirect fashion at some uncertain time influence the price at

which stock is sold.

Under these rules, the law could function to allow injunc-

tions against corrupt stock transactions before they occurred,

which is why in its initial encounter with the law in Federated

the New York Court of Appeals gave it such a full-throated en-

dorsement. But the situation becomes more nuanced in those

cases where the government seeks disgorgement of profits,

and injunction against future conduct or prosecution of possi-

ble criminal conduct. In these cases, it turns out that, both for

political and legal reasons, the question of scienter is injected,

even if it is formally irrelevant under the basic liability test. It is

therefore instructive to note that while Greenberg and Smith

fought ferociously against the Martin Act counts, in the end

they lost because they admitted to scienter even as they sought

to deny fraud, which in most instances is established by proof

of scienter. Similarly, the loud #ExxonKnew campaign

floundered precisely because there was no scienter that has

been established based on Exxon’s early investigations into

global warming or in the effort to prove that ExxonMobil is-

sued false statements because they ignored the impact of po-

tential climate regulation on future profits. ExxonMobil did

revise its reserves, not because of anything to do with potential

regulatory risks of global warming, but because of the huge

downward pressure exerted on oil prices from the momentous

technological innovations most intimately associated with ad-

vanced fracking techniques.

The most important lesson to learn from this account of

the Martin Act is that the quality of its public administration

depends not only on matters of institutional design, or on its

detailed substantive provisions, but perhaps even more on the

quality of the public officials who administer it. If they are

overconfident in their judgments, they will be overzealous in

their behavior, which will lead to overheated public rhetoric

on the one side and misguided legal results on the other.

There is in public life no set of procedural safeguards that can

2018] REGULATORY ENFORCEMENT UNDER THE MARTIN ACT 921

wholly protect against overly aggressive public officials—which

is not to say that safeguards should be disregarded. The pro-

tection against abusive discovery, for example, could be critical

in controlling excessive enforcement at either the state or fed-

eral level. Fortunately, the line seems to have held against

Schneiderman’s direct global warming charges against Exx-

onMobil. But we should not rejoice too much over his strategic

retreat. The public nuisance cases remain. Next time we may

not be so fortunate.