Why Mixed Motives Shouldn’t Really Matter in Insider Trading Law: A Reply to Andrew Verstein’s Mixed Motives Insider Trading


Andrew Verstein’s, Mixed Motives Insider Trading,1 is a fantastic article. It’s smart, well-written, and represents a real contribution to the insider trading literature while at the same time tying that literature to a broader conversation on the proper treatment of mixed motives spanning many different areas of the law. If any aspiring or junior law professor (or anyone, really) is on the lookout for models of excellent legal scholarship, look no further than this article. It’s the real McCoy.

This isn’t to say I agree with Verstein’s ultimate conclusion. In fact, I don’t. Specifically, I think his model overlooks an important scenario—where neither the permitted nor prohibited reasons for trading are on their own sufficient to compel a trade, but together they are. I believe, and demonstrate below, that when expanded to take into account this overlooked scenario, the resulting model actually argues for an awareness test for determining an insider trader’s requisite mental state,2 which is different from the primary motive test that Verstein is advocating for.

But more generally, Verstein and I disagree on the appropriate methodology for determining the correct mental state required for insider trading liability. He believes the right way to answer that question is through a detailed, top-down policy analysis. By contrast, I think the logic of the doctrine itself answers the question for us, and that answer is that insider trading liability should turn more on whether the insider trader is aware of material non-public information when they trade regardless of whether such information is not the only, or not even the primary, reason motiving the trade. In other words, I don’t think that whether one’s insider trading3 motives are mixed or not should really matter for determining liability.

Despite these disagreements, though, I think this article is a fantastic example of how to do smart doctrinal scholarship, and it should be one of the handful of articles that every scholar must contend with in analyzing the issue of the appropriate mental state required for someone to be guilty of violating insider trading law. In Part I, I provide an overview of Verstein’s article, praising its ingenuity while also noting a possible oversight that potentially undermines his ultimate doctrinal prescription. In Part II, I sketch my preferred approach to answering the question of the appropriate mental state for insider trading liability, which is rooted in my view of insider trading as facilitating a type of contracting for extra-compensatory damages to deter a hard-to-detect breach. Under that view of insider trading law, the correct approach to the mental state question is to simply look at when, under state law, insider trading violates a particular fiduciary or contractual duty. In my view, this approach argues in favor of an awareness test, not Verstein’s primary motive test.

An Ingenious Model But With a Possible Oversight

In his article, Verstein is interested in what is usually referred to as the scienter question in insider trading law.4 What is that question exactly? Well, for various reasons, U.S. insider trading law consists of federal common law that the Supreme Court has constructed on top of an anti-fraud statute.5 So, if insider trading is going to give rise to liability, it must constitute fraud, and fraud typically requires a particular state of mind, which is referred to as scienter. In the insider trading law context, this question has to do with the state of mind the insider has with respect to the information that underlies the insider trading.

Verstein’s Approach: A Top-Down Policy Analysis

Drawing on his previous work,6 Verstein makes the observation that an insider trader might have mixed motives. He might trade while in possession of both prohibited information (i.e., “this drug candidate is going to flop”) and permitted information (i.e., “I need money to pay for my daughter’s college”), and the question is how the law should treat such mixed motives. Typically, the law has taken two approaches with respect to the scienter question. Some courts adopt a “use test,” requiring that the insider actually use the prohibited reasons when deciding to trade.7 In that case, the insider who sells stock to pay for his daughter’s graduation escapes liability despite his knowledge of the failing drug candidate. Other courts, by contrast, favor an “awareness test” requiring nothing more than the insider be aware of the prohibited information,8 in which case our trader is in trouble even if they didn’t use that information in determining whether to conduct the sale.

Verstein takes a different approach, adopting a “primary motive” test, which “permits an action undertaken with two motives so long as the lawful motive predominates; if the unlawful motive predominates then the act is unlawful.”9 Under this test, unlike the awareness test, just because a trader discovers, by virtue of their fiduciary position, that a drug candidate is going to fail the FDA’s test doesn’t mean that they can’t sell their stock to pay for their daughter’s graduation. But unlike the use test, the mere fact that they have a mixed motive alone doesn’t defeat liability under the primary motive test either. The question under the primary motive test is whether the permitted motive outweighs the proscribed one.

The appeal of this test is that it satisfies what Verstein calls “[t]he equal profits principle,” the idea that “traders alike in all respects except for their knowledge of proscribed information should enjoy the same expected profits from trading.”10 In articulating this principle, Verstein’s concern is that the possession of proscribed information shouldn’t make the insider trader worse, or better, off than someone who lacks such information. So, if two firms, A and B, have made considerable investments to obtain lawful information about a company supporting a purchase (or sale) of that company’s stock but firm B happens to acquire some unlawful information in addition that also supports a purchase (or sale), that unlawful information shouldn’t prevent firm B from acting like firm A with respect to the lawful information.

If this equal profits principle is the operating principle that should animate the scienter question, and Verstein makes plausible arguments that it’s a good one,11 then we need to think about the type of test that best satisfies that principle. The awareness test might seem too restrictive based on our example because it would prevent firm B from trading like firm A, the firm with no unlawful information. The use standard by contrast does a better job because it achieves the result we’re looking for: Under the use standard, there would be nothing wrong with firm B going through with the purchase (or sale) just like Firm A. For this reason, we might be inclined to think that the use standard is preferable.

The problem however is that there might be other scenarios where the use standard doesn’t do as good of a job at tracking our baseline case where the trader lacks access to unlawful information. For example, what if we allowed (as Verstein does) the information to vary in terms of how compelling it is in justifying a particular trade? For example, let’s say that Firms C and D both have lawful information that on its own is sufficiently compelling to justify the sale of a particular stock, but Firm D additionally has unlawful information that would justify the opposite action, a purchase, and the unlawful information is even more compelling than the lawful information? In that scenario, under a use standard, Firm D would just abstain from trading because the unlawful information forecloses a purchase and also presumably neutralizes the lawful information that favored a sale. By contrast, the trader in the baseline case, Firm C in our example, would sell.

Because one is unlikely to come up with a workable test that causes the insider trader’s behavior to track, in every possible scenario, the behavior of a trader lacking unlawful information, it is necessary to make assumptions about expected trading profits and then figure out which rule gets us closest to the expected profits of that baseline case. This is precisely what Verstein does.12 He models mixed motives insider trading by assuming that reasons for trading are either lawful or unlawful and allows those reasons to vary in terms of strength. Importantly, however, Verstein assumes that in his baseline case, a given reason, whether permitted or prohibited, is always sufficiently strong to result in some trade.13 I think this assumption overlooks an important consideration that biases Verstein’s results. But more on this below.14 The point is that he ends up with eight trading scenarios where the information compels the trader to make a trading decision (purchase, sell, or simply abstain) and where a given type of information (lawful or unlawful) is allowed to vary in terms of how compelling it is in influencing the trading decision relative to the other type of information.15 Thus, the first two scenarios (domains I and II in Verstein’s model) are where both prohibited and permitted information support a purchase but the scenarios differ over which type of information is more compelling; the fourth and fifth scenarios (domains V and VI) are where both prohibited and permitted information support a sale and again the scenarios differ over which type of information is more compelling; and the remaining scenarios cover the mixed motive scenarios where permitted information supports a purchase (or sale), the prohibited information points in the opposite direction, and one signal is more compelling than the other (in domains III and IV, the prohibited information is more compelling; in domains VII and VIII, it’s the permitted information that has that honor).

Given this model, Verstein then assigns different expected values to these eight trading scenarios based on several plausible assumptions, including that the future stock price is correlated with unlawful reasons to trade.16 He considers two different types of insider traders: those whose legal reasons to trade are based on lawful information (“informational traders”) and those he calls “idiosyncratic traders” whose legal reasons to trade are idiosyncratic (like paying for a child’s college), the primary difference between the two types of traders being different expected payoffs.17 He demonstrates that with respect to the idiosyncratic traders, the primary motive test satisfies the equal profits principle.18 In other words, the primary motive test generates the same expected profits for the idiosyncratic trader with unlawful reasons as the same trader without such unlawful reasons. The case of informational traders is a bit more complicated. There, the primary motive test generates expected profits that are greater than those generated by the awareness test but still less than those generated by our baseline scenario of an informational trader who trades without unlawful information. The use test has the opposite problem—it generates greater expected profits than the baseline scenario. But Verstein argues that faced with such a choice, it’s preferable to follow a “less-than-equal profits principle” to avoid creating incentives to seek out unlawful information they might be able to shield with lawful motives.19

A Possible Oversight?

I should pause at this point to return to an objection I only mentioned in passing but that it is now necessary to revisit. There is obviously a limit to how many variables one can include in a model, particularly an informal one included in a law review format. However, I do think that Verstein overlooks an important scenario that, even under his own model’s assumptions, makes the awareness test look a whole lot more attractive than it does in his article.

The scenario I have in mind is where Firm A and B both have lawful information that alone isn’t sufficiently compelling to buy (or sell) the stock in question, but Firm B additionally has unlawful information that, when combined with the less-than-compelling lawful information, does provide sufficient reason to make the trade. Let us assume that the lawful information is relatively more important to the ultimate decision than the unlawful information, but the decision would not be made without the unlawful information.20 In that case, Firm A, our baseline trader who only trades on lawful information, would not trade. Under the awareness test, Firm B similarly would not trade. But under the primary motive test, Firm B would trade.

We can actually incorporate this “partial information” scenario fairly easily into Verstein’s model. It only pertains to those situations where the lawful and unlawful information provide the same trading signal, either a purchase or sale. In Verstein’s model, these are what he refers to as domains I, II, V and VI. I create four new domains, IA, IIA, VA and VIA. Domains I, II, V and VI mean exactly the same thing as Verstein means in his article, which is to say a mixed motives situation where each type of information, lawful and unlawful is, on its own, sufficient to compel a trade, but one type of information provides a greater trading signal than the other. These four additional domains—IA, IIA, VA and VIA—signify a situation where neither type of information, lawful or unlawful, provides a sufficiently strong signal to compel a trade, but together they do. Consider for example, domain IA, which refers to a situation where neither the lawful nor unlawful information is alone sufficient to justify a trade, but together they are—they signal a “buy”—and the lawful information is stronger than the unlawful information. Domain IIA is the same but where the unlawful partial information is stronger than the lawful partial information. Domains VA and VIA are like Domains IA and IIA, respectively, regarding the relative intensity of the lawful information compared to the unlawful,21 but with respect to a “sell” rather than “buy” signal. Here is how we might summarize the trading behaviors under the baseline scenario (what Verstein calls the “naïve trader”) under the use standard, awareness standard, and primary motive standard:



In Verstein’s model, which omits these partial information scenarios, the naïve trader never abstains from trading because the lawful information is always sufficient to compel a trade.22 That is no longer the case under our partial information scenarios, and so the naïve trader now abstains from trading in scenarios IA, IIA, VA and VIA because the naïve trader only acts on lawful information and the lawful information in those cases is partial—it is not sufficient to compel a trade. However, the primary motive test would allow a trade in domains IA and VA because in those cases, the lawful information is the primary motive even though it is not on its own sufficient. Here are the trading profits for idiosyncratic trading and informed trading23:

Notice that the results here are quite different from Verstein’s. Now, the primary motive test does not look as compelling as before. In both the idiosyncratic trading and the informed trading cases, the primary motive test generates greater expected profits than the baseline scenario of our “naïve trader.” Of course, in the idiosyncratic trading case, so does the awareness test. But importantly, those excess profits are less than under the primary motive test. And in the case of informed trading, the awareness test generates less expected profits than the baseline scenario. I would think that based on Verstein’s arguments for a less-than-equal profits principle as a sort of second-best case, he would have to concede that the awareness test is preferable than the primary motive and use tests in the case of informed trading. And I would think that the spirit of that same principle would also favor the awareness test in the case of idiosyncratic trading (the idea being that if no test satisfies the less-than-equal profits principle, choose the one that generates the smallest excess profits).

Now, Verstein no doubt has perfectly good responses to this objection about the incompleteness of his model. Perhaps he doesn’t think this scenario I identify—where lawful and unlawful information only compel a trade when combined—is actually as common as I suspect it might be or at least not as common as the scenario he focuses on where the one type of information is sufficient on its own to compel a trade. In that case, he might choose different probabilities that would affect the calculation of expected profits. Or maybe he would respond by modifying his primary motive test so that it’s a “primary sufficient motive” test, meaning that the primary motive also must be sufficient in and of itself to motivate the trade. However, if his goal is to cast a wider liability net than the use standard,24 I am not exactly sure that a primary sufficient motive test does the trick, considering how difficult it would be to prove a primary sufficient motive.

In other words, I believe that Verstein’s model is underspecified. It leaves out some important variables. And for this reason, I do not agree with Verstein’s ultimate conclusion even if I were willing to subscribe to his methodology of choosing the scienter test that comes closest to satisfying the equal profits rule. Nevertheless, I hope this criticism doesn’t cause one to overlook the impressive creativity of the presentation of Verstein’s argument. It is not easy in the law review format to illustrate the effects of rule changes without formal mathematical modeling. Verstein’s approach is excellent with frankly ingenious diagrams and a numerical overlay that does a good job illustrating his argument.

Where The Disagreement Really Lies

In other words, I really like Verstein’s article even though I think his model is under specified. But in the final analysis, I do not think that insider trading law should adopt Verstein’s primary motive test, for reasons in addition to, and independent from, the objection I raised in the previous section. I simply do not think that the proper approach to answering the scienter question is a “top-down” policy analysis divorced from doctrine that asks which test about one’s mental state comes closest to satisfying the equal profits rule. Rather, I think the logic of the doctrine, properly understood, itself tells us what the test should be, and it is not a primary motive test but a type of awareness test, or what one might call a “partial motive test.”25 This “bottom-up” approach strikes me as a better methodology for answering the scienter question.

My Approach: A Bottom-Up Approach

What do I mean when I say the logic of the doctrine itself tells us what the test should be? One can readily imagine certain types of fraud where the logic of the doctrine does not really tell us anything about the type of scienter test that should apply. Take, for example, a case involving an affirmative misstatement. If the CEO of a pharmaceutical company announces that they are confident that a current drug candidate will pan out and then it doesn’t, what’s the right scienter test for proving fraud? Is it recklessness, in which case evidence of the CEO’s knowledge of negative trial results might be sufficient to prove scienter? Or is it something else, like intent, in which case more evidence would be needed, for example, motive evidence like the CEO’s desire to keep the stock high in order to be able to close an all-stock acquisition of a competitor? The point is that there is no principle within the doctrine itself that tells us which test is the correct one.26

Insider trading law, however, is different. This is because insider trading liability requires the violation of a duty (supplied by either fiduciary law or an underlying contract), plus non-disclosure of that violation.27 As the Supreme Court has made clear, it is the non-disclosure, not the violation of the underlying duty, that makes the conduct fraudulent and therefore actionable under Rule 10b–5.28 But the violation of an underlying duty is nevertheless necessary, and to determine whether a violation has occurred, one must look at fiduciary duty law or the underlying contractual language, as the case may be.

When Does Insider Trading Violate Fiduciary Duties?

In Delaware, the question of when insider trading violates fiduciary duties is a matter of settled law.29 The Delaware courts have held that insider trading violates the duty of loyalty when material non-public information provides any motivation whatsoever for a fiduciary to trade on that information.30 This means that possession of material non-public information gives rise to an inference of a violation of the duty of loyalty, but that inference can be overcome if the defendant can show it did not actually use the information in the trade.31 Thus, as far as the Delaware courts are concerned, the mixed motive issue is resolved in favor of what one might call a partial motive test.32 Satisfaction of that test means the insider trading at issue constitutes a breach of the duty of loyalty.

To be clear, that fiduciary breach alone doesn’t constitute fraud for purposes of Rule 10b–5.33 However, once one proves the breach of a duty, the only thing left is to show non-disclosure of that breach. In other words, the mixed motives question no longer enters the picture. To summarize, then, when it comes to a case where the duty breached by the insider trading is supplied by fiduciary duty law, the relevant test is a partial motive test—the defendant can only avoid liability by demonstrating that they weren’t motivated in any way by prohibited information. In practice, this will undoubtedly look very similar to an awareness test.34

When Does Insider Trading Violate Contract Law?

That’s how the doctrine works when it comes to the breach of a fiduciary duty, but what about where the underlying duty is contractual, not fiduciary, in nature? For example, consider the facts of the well-known case involving Mark Cuban: Mark Cuban sold his shares in after learning from the CEO that Cuban’s ownership would be substantially diluted by a Private Investment in Public Equity (“PIPE”) transaction undertaken to recapitalize the faltering business.35 The U.S. Securities and Exchange Commission (“SEC”) argued that Cuban’s trade violated Rule 10b–5 because it was made in breach of a pre-existing confidentiality agreement, and the breach was concealed by Cuban.36 The court disagreed with the SEC, but not because of some extra-contractual principal about scienter like the one Verstein proposes in his article. Rather, the court disagreed because, in inspecting the contract, it didn’t think that the trade actually violated the contract’s terms, which barred the disclosure, not the use, of the non-public information.37 If the contract had also included a non-use provision, matters would have stood differently.

But of course, with respect to the scienter issue, the question remains: What precisely constitutes a violation of a non-use agreement? Under Delaware law, the answer is possession of the material information combined with a prohibited use.38 In other words, an awareness test. So, for example, in the Cuban case, if the court had found that Cuban’s trading had violated the confidentiality agreement, it would not have been an effective defense for Cuban to argue that the information nevertheless didn’t motivate the sale of his stock. If instead Cuban had been bound not by a confidentiality agreement but by fiduciary duties (for example, if they had been an executive at the company), evidence that the information didn’t motivate his trade would have been sufficient to defeat a claim of fiduciary breach.39 That’s the difference between the appropriate test when the insider trading is a breach of contract versus a breach of fiduciary duty. However, given how difficult it will usually be to prove that the information didn’t motivate the trade in any way, I think that the test for a fiduciary breach will generally be quite close to the awareness test applicable to a contract breach.

Thus, regardless of whether the insider trading breaches a fiduciary duty or a contractual one, the scienter test is basically an awareness standard.40 Of course, one might object by saying that, with respect to contractual duties, I haven’t actually shown that the scienter standard is an awareness standard. Rather, I’ve simply shown that in determining the appropriate standard, one must look to the contract in question. And that’s true, but non-use agreements are always drafted with an awareness standard. A non-use agreement prohibits putting covered information to prohibited uses, full stop.41 There might be various carveouts for certain types of information or for certain uses.42 But there’s never a carveout for the degree to which the covered information motivated a prohibited use. Could a non-use agreement take such an approach? Of course it could, and in that case, the court should defer to that language. But as a general matter, non-use agreements, like fiduciary duty law, require nothing more than awareness in determining whether insider trading violates the duties created by these sources of law.

To summarize then, in my view, insider trading law requires the violation of a duty, either fiduciary or contractual, plus non-disclosure of that violation. The scienter requirement is supplied simply by considering what constitutes the violation of the duty in question, and one does this by looking to state fiduciary duty law or the language of the contract in question, depending on the nature of the law that provides the underlying duty. One might refer to this as a “bottom-up” approach to solving the mixed motives problem because the scienter question is resolved by going to the source of the law, fiduciary or contractual, and asking what the test is for determining whether the duties supplied by those sources of law are violated by insider trading.

Why This Bottom-Up Approach?

But why is the bottom-up approach the right one? Sure, insider trading law might require the breach of a duty, plus non-disclosure of that breach. But we are, after all, talking about fraud here, and maybe plaintiffs should be required to prove more when we’re talking about insider trading as fraud than when we’re talking about insider trading as a breach of the duty of loyalty under Brophy or a breach of a confidentiality agreement as in Cuban.

There are two responses to this objection. First, plaintiffs do have to prove more when we’re talking about insider trading as fraud under Rule 10b–5—they have to prove non-disclosure as well. That’s the thing that separates fraud under Rule 10b–5 on the one hand and a mere breach of contract or fiduciary duty on the other.

The second response is that this bottom-up approach is justified in light of the unusual type of fraud at issue in insider trading law. As I’ve explained elsewhere,43 the best defense of our fraud-based insider trading regime is as an attempt to allow parties to contract for fraud-based remedies (in other words, extra-compensatory damages) to regulate their information sharing. Why is this necessary? Because some breaches are harder to detect than others, and using one’s information to profit from insider trading is one of these hard-to-detect breaches.44 In theory, the solution to the problem of detecting such a breach is actually fairly straightforward: Contractually require the party receiving the information to simply tell you if they’ve breached their fiduciary or contractual duties to not use the information for insider trading purposes.45 But that solution will only work if the breach of that disclosure obligation triggers extra-compensatory damages. After all, the insider trading itself already breaches the underlying covenant (supplied by fiduciary law or contract, as the case may be) that prohibits use of the information for insider trading purposes. In other words, the breaching party already owes compensatory damages for the insider trading before even considering the disclosure requirement.46 What incentive is there then for that breaching party to comply with the disclosure requirement if doing so won’t avoid some additional cost?

One might think that this incentive issue is easily overcome since usually non-disclosure in the face of a duty to disclose does warrant extra-compensatory damages under the theory of fraudulent non-disclosure.47 But remember that the disclosure requirement we’re talking about here is contractual in nature: It’s a contractual requirement to disclose the breach of an underlying fiduciary or contractual obligation prohibiting the use of information for insider trading purposes. And courts are generally very reluctant to award extra-compensatory damages in the case of contractually created non-disclosure provisions for fear of blurring the line between contract and tort law.48

As I have argued elsewhere, Rule 10b–5 should be understood to have changed all of this, at least with respect to the very specific hard-to-detect breach of insider trading.49 When it comes to insider trading, the argument is that we should allow parties to contract for fraud damages by way of contractual provisions requiring disclosure of a breach of fiduciary or contractual prohibitions on the use of information for insider trading purposes. Furthermore, we should find that such disclosure requirements are simply implied in certain circumstances, including fiduciary relationships, where, by a hypothetical bargain analysis, we can reason that parties are likely to have agreed to such disclosure requirements if they had specifically addressed it. The result of this analysis is that we end up with an insider trading doctrine that looks an awful lot like the one we have: the requirement of the breach of an underlying duty, whether fiduciary or contractual, plus nondisclosure of that breach.50

Given this understanding of insider trading law, the bottom-up approach for settling the scienter question makes sense. Requiring something more than what is necessary to establish a breach of the underlying duty (for example, Verstein’s primary motive or even a traditional “sole motive” use test) would disrupt this underlying policy allowing parties to contract for extra-compensatory damages to protect their information from being misappropriated for insider trading purposes.

Why Verstein Rejects the Bottom-Up Approach

To be clear, Verstein doesn’t entirely overlook the bottom-up approach to answering the awareness/use question, which I’ve adopted here.51 But I do think that he rejects it too quickly and therefore fails to appreciate its force. His real target is the argument, explored by Donna Nagy, among others,52 that the scienter requirement should be derived from underlying fiduciary principles.53 Verstein points out that doing so would not actually resolve the scienter question in every case because not every case of insider trading liability involves the breach of a fiduciary duty.54 What Verstein has in mind are cases involving “classical insiders,” which is to say, situations where the insider trader is trading on material non-public information derived from his employer. In those cases, courts typically apply what is known as the “classical theory,” which requires a fiduciary or fiduciary-like relationship with the counterparty to the trade,55 as opposed to the “misappropriation theory,” which requires that such a relationship exist with the source of the information.56 While a fiduciary relationship is said to exist between the classical insider and the stockholders of the insider’s company,57 the same can’t be said of the relationship between the insider and his company’s bondholders or a competitor’s stockholders. Thus, to extend liability under the classical theory to the classical insider who trades in his company’s bonds or a competitor’s stock, one would need to focus on some non-fiduciary “relationship of trust and confidence” with the counterparties to those trades. It is presumably for this reason that Verstein remarks that looking to fiduciary principles to resolve the scienter requirement will fail because “[it] would, at best, only resolve half of it.”58

But this doesn’t serve as an objection to the bottom-up approach for at least two reasons. First, even if that bottom-up approach’s applicability depended on the insider trading in question violating an underlying fiduciary relationship (as opposed to a non-fiduciary relationship of trust and confidence), the examples Verstein cites can easily be reframed to involve such fiduciary relationships. It’s true that the federal courts tend to apply the classical theory to cases involving classical insiders and the misappropriation theory to everything else.59 And it’s also true that following that trend, there might be no relevant fiduciary breach in the case of the CEO trading in their company’s bonds or in a competitor’s stock. But, as I’ve argued elsewhere, there’s no reason to artificially constrain the doctrine’s applicability in this way.60

If one were willing instead to apply the misappropriation theory to cases involving classical insiders, many if not most insider trading cases that don’t seem to involve fiduciary breaches under the classical theory suddenly do under the misappropriation theory. To take Verstein’s example, if the CEO is trading in their company’s bonds or in a competitor’s stock, it’s probably because of information that they acquired from their own company by virtue of their position as CEO. And technically, fiduciary duties are owed to the corporate entity.61 Thus, under the misappropriation theory, which requires the breach of a duty owed to the source of the information, the CEO in those examples would be trading in breach of fiduciary duties owed to their company.62

 So, even if the bottom-up approach for determining the correct scienter standard required the insider trading to violate a fiduciary relationship, most insider trading cases do involve such relationships if reframed under the misappropriation theory. Of course, this isn’t true of every insider trading case. SEC v. Cuban, discussed earlier, is one such case where the only duty breached, if any, is contractual in nature. But that just brings me to my second reason why Verstein’s objection doesn’t address the bottom-up approach I’m arguing for here: That approach doesn’t require us to derive the correct scienter standard exclusively from fiduciary principles. Rather, under the bottom-up approach, the scienter question boils down to whether there is a violation of a duty, which, depending on the legal source of the underlying duty, is a question for state fiduciary duty law or the language of the underlying contract. The answer to that question is basically an awareness test, not the primary motive test advocated for by Verstein.


Verstein and I disagree on the question of the proper scienter rule in insider trading law. He argues for a primary motive test based on a top-down policy approach that I would argue overlooks an important consideration that undercuts his policy prescription. I prefer an awareness test but for a different reason entirely: because I prefer a bottom-up approach where the answer to the scienter question is supplied by the relevant law, either fiduciary or contractual, that gives rise to the duty breached by the insider trading. The difference is that I believe the best defense of insider trading as a type of fraud is one that views Rule 10b–5 as an attempt to allow parties in an information-sharing relationship, whether fiduciary or otherwise, to do something the common law historically forbade, which is to contract for extra-compensatory damages for preventing the illicit use of confidential information. Given this understanding, the scienter requirement should be derived from that underlying information-sharing relationship, whether fiduciary or otherwise. Verstein no doubt has an entirely different view of the purposes of insider trading law. But despite these differences, it is always a pleasure to spend time reflecting on Verstein’s thinking, and Mixed Motives Insider Trading is a phenomenal article that will enrich the conversation about insider trading law doctrine for years to come.

  1. [1]. Andrew Verstein, Mixed Motives Insider Trading, 106 Iowa L. Rev. 1253 (2021).

  2. [2]. In this Response, I use the term “insider trader” as someone whose trading in securities on the basis of material non-public information would be actionable under Rule 10b-5 if the applicable mental state requirement is met.

  3. [3]. In this Response, I use the term “insider trading” to refer to trading on the basis of material, non-public information. What it means to trade “on the basis of” such information is the issue addressed herein.

  4. [4]. See 18 Donald C. Langevoort, Insider Trading Regulation, Enforcement and Prevention § 3:13 (updated May 2021) (outlining case law in relation to the scienter element of insider trading). I realize that Verstein himself wants to stay away from framing the problem as one involving scienter, preferring instead to refer to it as the awareness/use problem or mixed motives problem. See Verstein, supra note 1, at 1268–69 (discussing how the “awareness/use debate” is sometimes focused on different elements of the insider trading offense, including that of scienter). I sometimes use Verstein’s preferred “awareness/use” nomenclature to refer to the issue in question. Other times, I use the term “scienter” simply because I think this is how most courts refer to this problem. See, e.g., Zachary J. Gubler, A Unified Theory of Insider Trading Law, 105 Geo. L.J. 1225, 1227 n.2 (2017). By doing so, I don’t mean to imply that the scienter elements somehow help address the issue as to where on the awareness/use spectrum the proper test lies. See Verstein, supra note 1, at 1268 (identifying problems in trying to argue that “the solution to the awareness/use debate may be implied by whatever element it is part of”).

  5. [5]. The anti-fraud statute is Section 10(b) of the Securities Exchange Act of 1934. See
    15 U.S.C. § 78j(b) (2018) (making it illegal “[t]o use or employ, in connection with the purchase or sale of any security registered on a national securities exchange or any security not so registered, or any securities-based swap agreement any manipulative or deceptive device or contrivance in contravention of such rules and regulations as the Commission may prescribe as necessary or appropriate in the public interest or for the protection of investors”) (footnote omitted). The “Commission” referenced in that provision is of course the Securities and Exchange Commission (“SEC”), which promulgated Rule 10b-5 in accordance with Section 10(b). See 17 C.F.R. § 240.10b–5 (2019).

  6. [6]. See generally Andrew Verstein, The Jurisprudence of Mixed Motives, 127 Yale L.J.
    1106 (2018) (discussing mixed motives for conduct in a variety of domains of law).

  7. [7]. See, e.g., United States v. Anderson, 533 F.3d 623, 629 (8th Cir. 2008).

  8. [8]. See, e.g., United States v. Teicher, 987 F.2d 112, 120–21 (2d Cir. 1993).

  9. [9]. Verstein, supra note 1, at 1298.

  10. [10]. Id. at 1279.

  11. [11]. See id. at 1278–83 (explaining the principle is one that is “intuitively plausible, and broadly compatible with all normative positions”). However, as discussed in Part II below, I think that we should take a different approach to determining the requisite mental state for insider trading liability.

  12. [12]. Id. at 1287–91.

  13. [13]. We can see this, for example, in Verstein’s Figure 8, where he shows that the “naïve trader”—the trader who only trades in line with his permitted reasons, in other words, the baseline scenario—always makes a trade. He never abstains from trading because those permitted reasons aren’t sufficiently compelling in and of themselves to compel a trade. Verstein, supra note 1, at 1292.

  14. [14]. See infra notes 20–24 and accompanying text.

  15. [15]. These eight trading scenarios represent all of the permutations generated from combining permitted and prohibited information, each of which signals either a “buy” or “sell” and where the signal is allowed to vary in terms of intensity. Thus, what Verstein calls “Domain I” is where the permitted and prohibited information both signal a “buy” but the permitted information is more compelling than the prohibited; Domain II is where both types of information also signal a “buy” but the prohibited information is more compelling than the permitted; Domain III is where both types of information signal a “sell” but the permitted information is more compelling than the prohibited; and so on. Verstein, supra note 1, at 1292.

  16. [16]. See id. at 1288.

  17. [17]. See id. at 1289–90.

  18. [18]. See id. at 1302.

  19. [19]. See id. at 1303.

  20. [20]. For example, the lawful information might get Firm B 45 percent of the way there, but the unlawful information provides the remaining 5 percent needed to compel the firm to trade.

  21. [21]. In other words, like Domain IA, in Domain VA, the permitted information creates a stronger signal than the prohibited information, and like Domain IIA, in Domain VIA, the prohibited information is the stronger signal.

  22. [22]. See Verstein, supra note 1, at 1295.

  23. [23]. I am assuming that this partial information scenario is as likely as Verstein’s assumption of a sufficient information scenario and therefore I just divide the probabilities in half with respect to the affected domains. So, for example, in the case of idiosyncratic trading, Verstein assumes that there is a 16 percent chance of being in domain I, where the permitted reason for trading is stronger than the proscribed reason. In my modified model, there is now only an 8 percent chance of being in Domain 1 (half of 16 percent) and another 8 percent chance of being in domain IA where the permitted reason for trading is also stronger than the proscribed reason but is not on its own sufficient to trade. I apply this same convention to the other affected domains, in other words domains II, V and VI, as well as to the case of informed trading.

  24. [24]. See Verstein, supra note 1, at 1303 (observing that the use test “promises too much profit to informed traders” and rejecting it over the primary motive test for this reason among others).

  25. [25]. I am equivocating slightly here simply because I think the test differs, albeit subtly, depending on whether the underlying duty comes from fiduciary or contract law. See infra notes 29–42 and accompanying text.

  26. [26]. The weight of authority among the federal courts is that recklessness is sufficient in this scenario. See, e.g., Sanders v. John Nuveen & Co., 554 F.2d 790, 793 (7th Cir. 1977).

  27. [27]. See United States v. O’Hagan, 521 U.S. 642, 651–56 (1997).

  28. [28]. See id. at 660. Rule l0b–5 makes it unlawful, among other things, “[t]o employ any device, scheme, or artifice to defraud, [or] ... [t]o engage in any act, practice, or course of business which operates or would operate as a fraud or deceit upon any person.” 17 C.F.R.
    § 240.10b–5(a), (c) (2019).

  29. [29]. See In re Oracle Corp., Derivative Litigation, 867 A.2d 904, 933 (Del. Ch. 2004) aff’d, 872 A.2d 960 (Del. 2005).

  30. [30]. See id. at 934 (“Distilled to its essence, therefore, a plaintiff seeking to prevail on a Brophy claim ultimately must show that: 1) the corporate fiduciary possessed material, nonpublic company information; and 2) the corporate fiduciary used that information improperly by making trades because she was motivated, in whole or in part, by the substance of that information.”). It should be noted that then-Chancellor Strine mentions in passing that he understands this standard to be in line with the scienter requirement for federal insider trading law, but that does not appear to be the reason he is articulating the standard in the way he does. Rather, he is principally focused on the state law precedent. So, I do not think this is susceptible to the charge that he is just basing this off of federal scienter requirements and thus it is circular to look to the state law.

  31. [31]. See generally Kahn v. Kolberg Kravis Roberts & Co., 23 A.3d 831 (Del. 2011) (approving a special litigation committee’s analysis, finding that, despite possessing material non-public information regarding a planned asset sale, board members did not violate their duty of loyalty by purchasing preferred stock because that information did not motivate their decision to undertake the trade).

  32. [32]. Interestingly, the SEC has at times endorsed such a test for insider trading in violation of Rule 10b–5. See In re Investors Management Co., 44 S.E.C. 633, 1971 WL 120502, at *6 (July 29, 1971) (concluding that insider trading liability required “that the information be a factor in [the insider’s] decision to effect the transaction”).

  33. [33]. See Santa Fe Indus., Inc. v. Green, 430 U.S. 462, 472–74 (1977).

  34. [34]. Under a true “awareness” test, the defendant would presumably not avoid liability even if he could show that he wasn’t motivated by the unlawful information in any way, as long as he was simply aware of it. However, in practice, awareness likely gives rise to an inference of motive, and therefore an awareness test looks a lot like a partial motive test.

  35. [35]. SEC v. Cuban, 620 F.3d 551, 557–58 (5th Cir. 2010).

  36. [36]. See id. at 552–53.

  37. [37]. The Fifth Circuit remanded the case for determining whether Cuban had indeed agreed to a non-use provision. See id. at 558. On remand, the jury found this not to be the case. See Natalie Posgate & Mark Curriden, Jury Finds No Proof Mark Cuban Engaged in Insider Trading, Dallas Morning News (Oct. 16, 2013, 10:47 PM),
    2013/10/16/jury-finds-no-proof-mark-cuban-engaged-in-insider-trading [

  38. [38]. See Martin Marietta Materials, Inc. v. Vulcan Materials Co., 68 A.3d 1208, 1218–19 (Del. 2012). There, the question was whether Martin Marietta breached a confidentiality agreement with Vulcan prohibiting it from using Vulcan’s information for purposes of a hostile bid for Vulcan. The question is not whether the information subject to the confidentiality agreement was the only information that motivated the hostile bid but rather that it was put to some prohibited use like a hostile bid. In other words, the question seems to be whether the hostile bid was motivated, in whole or in part, by the covered information. To defeat Martin Marietta’s claim, it wouldn’t have been enough for Vulcan to show that it had used other permissible information as well.

  39. [39]. See supra notes 29–34 and accompanying text.

  40. [40]. As discussed earlier, the scienter test where the insider trading breaches a fiduciary duty is probably best described as a partial motive test. But that test in practice probably operates like an awareness test. See supra note 34.

  41. [41]. See, e.g., ABA Mergers & Acquisitions Comm., Model Confidentiality Agreement in Model Merger Agreement for the Acquisition of a Public Company 341, 347, 350 (2011).

  42. [42]. See id. at 353.

  43. [43]. See Zachary J. Gubler, Insider Trading as Fraud, 98 N.C. L. Rev. 533, 537–42 (2020).

  44. [44]. For a discussion of how the costs of verifying a contract breach can cause contracting to break down, see Aaron S. Edlin & Alan Schwartz, Optimal Penalties in Contracts, 78 Chi.-Kent L. Rev. 33, 33–35 (2003); Gregory Klass, Contracting for Cooperation in Recovery, 117 Yale L.J. 2,
    5–9 (2007).

  45. [45]. See Klass, supra note 44, at 36–38, 42–49.

  46. [46]. This is of course true of the breach of a contract. But it is also generally true of fiduciary breaches as well. See generally Daniel Kelly, Double Deterrence: Restitution and Punitive Damages in Trust and Fiduciary Law, 107 Iowa L. Rev. (forthcoming 2022),
    pers.cfm?abstract_id=2952755 [] (discussing how traditionally punitive damages weren’t available for breaches of fiduciary duties).

  47. [47]. See, e.g., Restatement (Second) of Torts § 551(1)–(2)(a) (Am. L. Inst. 1977) (“(1) One who fails to disclose to another a fact that he knows may justifiably induce the other to act or refrain from acting in a business transaction is subject to the same liability to the other as though he had represented the nonexistence of the matter that he has failed to disclose, if, but only if, he is under a duty to the other to exercise reasonable care to disclose the matter in question. (2) One party to a business transaction is under a duty to exercise reasonable care to disclose to the other before the transaction is consummated, (a) matters known to him that the other is entitled to know because of a fiduciary or other similar relation of trust and confidence between them ... .”).

  48. [48]. See Klass, supra note 44, at 33–36.

  49. [49]. See Gubler, supra note 43, at 565–66.

  50. [50]. To be clear, the implication here is that while a court should find an implicit requirement in every fiduciary relationship to disclose the breach of the underlying duty, the court should not find such an implicit requirement in every single contractual relationship involving a confidentiality agreement. See id. at 574–80.

  51. [51]. See Verstein, supra note 1, at 1271–72.

  52. [52]. Donna M. Nagy, The “Possession vs. Use” Debate in the Context of Securities Trading by Traditional Insiders: Why Silence Can Never Be Golden, 67 U. Cin. L. Rev. 1129, 1160–62 (1999).

  53. [53]. See Verstein, supra note 1, at 1272.

  54. [54]. See id.

  55. [55]. See Chiarella v. United States, 445 U.S. 222, 227–29 (1980).

  56. [56]. See United States v. O’Hagan, 521 U.S. 642, 652 (1997).

  57. [57]. I say “is said to exist” because in actuality, directors and officers owe fiduciary duties to the corporate entity itself, not to individual shareholders. See, e.g., Goodwin v. Agassiz, 186 N.E. 659, 660 (Mass. 1933); Restatement (Second) of Agency § 14C cmt. a (Am. L. Inst. 1958) (stating that directors owe duties “to the corporation itself rather than to the shareholders individually or collectively”); Margaret M. Blair & Lynn A. Stout, A Team Production Theory of Corporate Law, 85 Va. L. Rev. 247, 293 (1999) (“In recent years, it has become common in both the legal and the economic literature for directors’ fiduciary obligations to be described as being owed ‘to shareholders.’ Yet case law makes clear that directors owe their fiduciary duties primarily to the corporation itself.”) (footnote omitted). Nevertheless, the Supreme Court ignored this reality in articulating the classical theory. See Chiarella, 445 U.S. at 227–28.

  58. [58]. Verstein, supra note 1, at 1272.

  59. [59]. See Gubler, supra note 4, at 1230 n.22.

  60. [60]. See id. at 1228–32.

  61. [61]. See supra note 57.

  62. [62]. See Gubler, supra note 4, at 1231.


Marie Selig Professor of Law, Arizona State University Sandra Day O’Connor College of Law.

All errors are mine.