New York Attorney General Overreaches in Climate-Change Complaint Against Exxon

The New York Attorney General (“NYAG”) recently filed a complaint against ExxonMobil alleging that the company violated the Martin Act, New York State’s securities law, by making material misstatements concerning how it was accounting for the possibility of increasingly stringent climate-change regulations in the future.[1]  Exxon publicly projected costs arising from the regulation of carbon emissions that, for OECD countries, would reach $60 per ton of CO2  by 2030 and $80 per ton by 2040, with similar regulations in non-OECD countries also starting to impose costs by 2030.  The complaint alleges that in making many of its investment decisions, Exxon did not assume that the costs of such regulation would be as high as its public projections. Nor, the complaint alleges, did Exxon assume regulatory costs this high in its determination of its published proved reserves and oil and gas resource base assessments or its published projections of the future demand for, and prices of, oil and gas, or in its determination of whether any assets should be publicly accounted for as impaired.

The complaint represents the latest chapter in a multi-year investigation of Exxon begun by former NYAG Eric Schneiderman.  Prompted by evidence that the company’s public climate change statements were at odds with what some of its own scientists were saying internally, Schneiderman subpoenaed a broad range of Exxon’s internal documents.  He did so utilizing his special powers under the Martin Act to subpoena private documents without either obtaining grand jury approval, which is required in most ordinary New York criminal proceedings, or the filing of a complaint, which is required in an ordinary civil action and is subject to court review.[2]  In a column I wrote two years ago in the National Law Journal, I criticized the subpoena served upon Exxon as an abuse of these powers.[3]  My concern was that no matter what the subpoena might turn up, it was very unlikely that the NYAG would be able to plausibly argue that Exxon had in fact violated the Martin Act by omitting mention of the views of its scientists.  This omission would only have been a Martin Act violation if it had made Exxon’s public statements concerning climate change materially misleading, which it almost certainly did not.  As discussed below, materiality requires a substantial likelihood that the omitted information will be viewed by the reasonable investor as having significantly altered the total mix of available information.  Given that Exxon had only a small fraction of all the world’s scientists working on climate change, it is unlikely that full disclosure of the Exxon scientists’ particular views on the subject would have changed the mind of the reasonable investor about whether to buy or sell Exxon shares.

The recently filed complaint goes off in a different direction, one involving alleged misstatements that more plausibly could be of importance to the reasonable investor.  Still, the suit faces a number of difficulties going forward and continues to raise troublesome questions of policy concerning the use of the Martin Act’s grant of subpoena powers.

Difficulties Going Forward

The Martin Act prohibits “any fraud, concealment, suppression [or] false pretense” in connection with the purchase or sale of a security.  It has been interpreted as banning, in connection with any such purchase or sale, “all deceitful practices contrary to the plain rules of common honesty,” including “all acts . . . which do by their tendency . . . deceive or mislead the public.”[4]  One way to commit such a violation, and the one relevant to this case, is for an issuer to publicly make an untrue statement of a material fact, or to omit to state a material fact necessary to make a public statement of the issuer not misleading.  Thus, there are two key questions: (1) does the complaint successfully allege that Exxon made one or more untrue or misleading statements, and (2) if it does, for each such misstatement, is the fact that is falsely stated or omitted material?

Did Exxon make any misstatements? The Exxon projections of the possible impact of future climate regulations in terms of dollars per ton of CO2 emissions (what it terms as its regulatory cost “proxies”) were made in a series of reports published between 2010 and 2014 (the “Reports”).  Exxon characterizes these proxies as “our effort to quantify what we believe government policies over the Outlook period could cost.”[5]

Publication of these proxies is not by itself the issue.[6]  Rather, the question of whether Exxon made a misstatement relates to what Exxon said or implied about its use of these proxies. Assume for purposes of discussion that Exxon, as alleged, in fact did not assume regulatory cost projections as high as the proxies when it made its investment decisions, its published assessments of its proved reserves and resource base, its decisions as to whether any assets should be publicly accounted for as impaired, and its published projections of the demand for, and prices of, gas and oil in the future.  One way Exxon could have made a misstatement would be to have explicitly stated somewhere that in undertaking these activities, it was using the proxies.  That would have been an untrue statement.  Determining whether Exxon actually made any such untrue statement will involve a careful review of everything that Exxon has said in its full range of public statements and SEC filings.

The other way Exxon could have made a misstatement would be if Exxon made a statement, in the Reports or elsewhere, that, while literally true, created an impression that would have misled a reasonable investor into believing that Exxon was using the proxy costs in undertaking these various activities.  In other words, the claim would be that Exxon omitted to state a fact (its failure to use the proxies as its projections of regulatory costs when undertaking these various activities) necessary to make the literally true statement not misleading.  Determining whether this was the case again requires a careful parsing of Exxon’s various public statements relating to the projections to see if there are any statements that standing by themselves, while literally true, lead to the impression that Exxon used the proxies in undertaking these various activities and that there were no other Exxon public statements that would have corrected for this misleading impression.

With regard to these two possible ways that Exxon might have made a misstatement, the complaint quotes numerous Exxon statements mentioning the proxy approach and alluding to the need to account for greenhouse gases, such as “we require that all our business segments include, where appropriate, GHG [greenhouse gas] costs in their economics when seeking funding for capital investments.”  Overall, though, the complaint is surprisingly vague in terms of identifying statements concerning the use of the proxy costs that are either explicitly false or, though literally true, are misleading standing alone and remain so even after considering everything else that Exxon has said.

If there were any misstatements, were they material? Assume further, as alleged, that Exxon made a misstatement to the effect that it used the proxy costs set out in the Reports in undertaking the various activities discussed above when,  in fact, it did not.  For each of these activities, is the fact that Exxon did not use the proxies material?  To be material, there must be a substantial likelihood that a reasonable investor would consider the misrepresented or omitted fact to be important in a decision to buy, hold, or sell Exxon shares.  Put another way, there needs to be a substantial likelihood that the omitted information would have been viewed by the reasonable investor as having significantly altered the total mix of available information.

First, consider investment decisions, which, for reasons discussed below, are probably the NYAG’s most promising candidate for establishing materiality.   Make the yet further assumption that because of the misstatement, the market believed that Exxon was using the proxies in its investment decisions when it was not.[7]  Four factors are key to determining whether the falsely stated or omitted fact was material: (i) the proxy costs stated in the Reports, (ii) the market’s mean expectation of the costs of future carbon emission regulation,  (iii) the costs of future carbon emissions regulation actually assumed by Exxon in making its investment decisions, and (iv) the sensitivity of Exxon’s level of investment to the level of future carbon emission regulation costs that it assumes and the impact on future cash flows to shareholders arising from an investment level determined using an incorrect assessment of these costs.

Suppose, for example, the market’s mean expected costs were as high as proxies.  If, unbeknownst to the market, Exxon determined its level of investment while assuming future regulatory costs lower than the proxies, Exxon would be using a flawed decision process resulting in overinvestment relative to what, in the market’s best estimate of these regulatory costs, would maximize the value of Exxon shares.  Whether or not knowledge of this flaw would be important to a reasonable investor, and hence be considered material, would depend on how great the difference was between Exxon’s actual cost assumptions and the proxy costs and, importantly, on the sensitivity of Exxon’s level of investment and future cash flows to the difference between the two.

A contrasting example has different implications, however.  Suppose the market’s mean expected costs were lower than the proxy costs but about equal to Exxon’s actual cost assumptions.  In that case, Exxon would in fact be investing at the level that would maximize the value of its shares, given the market’s best estimate of these regulatory costs, but the misstatement would result in the market incorrectly believing that Exxon was using a flawed decision process, one that this time would have led to underinvestment relative to what would maximize the value of its shares.  Again, the materiality of the misstatement would depend on how great the difference was and the sensitivity of the level of investment and future cash flows to such a difference.  If the misstatement was material, it could be a disclosure violation, just as in the first example.  It could not support the NYAG’s claim for damages for New York State employee pension funds that purchased Exxon stock, however, because the misstatement would have led the market to underprice, not overprice, Exxon shares.

Next consider Exxon’s published assessments of its proved reserves and resource base and its decisions as to whether any assets should be publicly accounted for as impaired.  If there were misstatements to the effect that Exxon used its proxies in making these assessments when it did not, a determination of their materiality would include consideration of factors similar to those discussed above with regard to investment decisions.  A further complication, however, is that these assessments are subject to strict accounting rules, compliance with which presumably would severely undermine the Finally consider Exxon’s published projections of the demand for, and prices of, oil and gas in the future.  If Exxon made misstatements to the effect that it used its proxies in making these assessments when it did not, there is little chance that they would be considered material. Given all the other publicly available information relevant to predicting the future demand and prices for oil and gas, Exxon’s projections are unlikely to have significantly altered the total mix. So even if Exxon’s projections were misstatements, they were not material misstatements.

Troublesome Questions Concerning the NYAG’s Subpoena Powers

Notwithstanding these difficulties, suppose that the NYAG is able to establish that Exxon made one or more material misstatements constituting violations of the Martin Act.  We would still be left with troubling questions concerning the way that the Exxon investigation began and Attorney General Schneiderman’s Martin Act subpoena that generated the needed evidence.

In public statements between 1997 and 2007, Exxon, while not denying the possibility that carbon emissions were causing warming, stressed the scientific uncertainties concerning the extent of carbon’s contribution to it and suggested caution in adopting costly rules curbing emissions before more was known.  Certain Exxon scientists, however, had apparently already internally urged the company to account for potential climate change in decisions concerning future oil and gas exploration and development, a fact that Exxon did not disclose.  From the beginning, however, whatever the subpoena might ultimately turn up, it was very difficult to see how the omission of the internal scientists’ views could be shown to be material – to significantly alter the total mix of publicly available information concerning carbon emissions’ contribution to climate change – and hence to have constituted a Martin Act violation.  Indeed, not surprisingly, the recently filed complaint is in no way based on the fact that Exxon made these statements concerning scientific uncertainty while omitting to disclose the internal views of its own scientists.

The Exxon subpoena may nominally have been about misled investors, but it was really about then-Attorney General Eric Schneiderman acting as a champion in the fight against global warming, a point upon which I elaborate in my column of two years ago.  In essence, a worthy end was being pursued by less than worthy means.  The experience with a number of prior NYAG Martin Act investigations seemed to provide insight into Schneiderman’s likely strategy in the Exxon investigation.  In these earlier investigations, the subpoena allowed an extensive search of the target’s corporate documents, some of which inevitably turned out to be embarrassing.  The investigation then ended with an “Assurance of Discontinuance,” whereby the target corporation agreed to undertake certain measures or changes in behavior.  The attorney general was able to secure this agreement without ever having to articulate a theory of how the facts available at the time the subpoena was issued could seriously suggest a violation of the Martin Act.  These investigations represented a troubling pattern: single official, without any legislative guidance or judicial oversight, was repeatedly able to use such a strategy to make public policy over almost any kind of business activity.

In the case of the Exxon investigation, past has turned out not to be prologue.  The subpoena has ultimately been followed not by an Assurance of Discontinuance, perhaps because Exxon simply refused to settle, but by the filing of a complaint against Exxon based on evidence quite distinct from the evidence justifying the subpoena.   Given the difficulties outlined above, Exxon may well defeat the action at the motion to dismiss or some later stage in the proceedings.  Still, the complaint at least alleges statements and actions that, if true, could, with further elaboration, plausibly establish that Exxon made one or more misstatements that would have been material to investors considering whether to buy or sell Exxon shares.  So it is possible that the NYAG could ultimately prevail.  What is troubling is how the evidence behind the complaint was obtained.  The subpoena power granted to the NYAG by the Martin Act was invoked based on evidence of alleged misstatements that could not have plausibly been of importance to the reasonable investor and hence to constitute Martin Act violations.

Eric Schneiderman’s successor, Barbara Underwood, inherited the Exxon investigation.  There is every reason to believe that she in good faith views the evidence cited in the complaint’s allegations as making out a colorable case for a Martin Act violation.  She also presumably believes that this evidence was gathered pursuant to a legally issued subpoena.  The problem again goes back to the original use of the Martin Act’s grant of subpoena power.  Where initially there is no serious evidence of a securities law violation, it is just as undesirable for this power to be used for a fishing expedition trying to come up with meaningful evidence of such a violation as it is for this power to be used to obtain evidence giving the NYAG leverage to obtain an Assurance of Discontinuance of some non-securities-related activity of which he or she disapproves. Thus the filing of this complaint simply adds to the argument that the NYAG should in the future be required, whether by the Martin Act’s amendment or mere public pressure, to provide a coherent explanation of how then available evidence suggests a real possibility of a genuine securities law violation before he or she uses these subpoena powers.


[1] People of the State of New York v. Exxon Mobil Corporation, Index No. 452044/2018 (hereinafter “NYAG Complaint”)

[2] N.Y. Gen. Bus Law Article 23-A, §352.

[3] See Merritt B. Fox, ExxonMobil is being Investigated, But Here’s the Real Problem, The National Law Journal, August 15, 2016.  Available at

[4] People v. Federated Radio, 244 N.Y. 33, 38-39 (1926).

[5] NYAG Complaint ¶92.

[6] Given all the other information available relevant to predicting the costs of future regulations, Exxon’s projections are unlikely to have significantly altered the total mix.

[7] This assumption cannot be taken for granted.  What market participants believe depends not only on the misstatement, but on all other public information that could bear on the question of what future regulatory cost assumptions Exxon used in making its investment decisions.  Suppose this other public information would lead the reasonable investor to believe that Exxon was using regulatory cost projections lower than the proxies when it was making its investment decisions. Then, if the basis for claiming that an Exxon statement to be a misstatement is that it is an untrue statement of fact, there may not be a substantial likelihood that if Exxon had instead disclosed the true situation, a reasonable investor would have regarded the total mix of information available as significantly altered, in which case the misstatement would not be considered material.  This is also the case  if the basis for claiming that the Exxon statement to be a misstatement is that it is misleading,  rather than that it is explicitly false.  Indeed, in this second situation, the statement might not even be regarded as misleading given this other public information.  A misleading statement is literally true.  What makes it misleading is that it leads the reasonable investor to assume a state of the world contrary to what the person making the statement knows to be true.  Thus, determining whether a literally true statement is nevertheless misleading depends on context.  The likelihood that a reasonable investor will be misled by the statement depends critically on what else is publicly known.

This post comes to us from Merritt B. Fox, the Michael E. Patterson Professor of Law at Columbia Law School, co-director of the Center for Law and Economic Studies. and co-director of the Program in the Law and Economics of Capital Markets.

1 Comment

Comments are closed.