More
    StartKryptowährung NewsCrypto Assets and Insider Trading Law's Domain

    Crypto Assets and Insider Trading Law’s Domain

    <?xml version=“1.0″ standalone=“yes“?><?xml version=“1.0″ standalone=“yes“??><?xml version=“1.0″ standalone=“yes“???><?xml version=“1.0″ standalone=“yes“????>

    I. Introduction

    An extensive literature addresses the substance of insider trading law. For example, should new techniques of high frequency trading be penalized as a species of “Insider Trading 2.0?”1 Should all insider trading be decriminalized?2 Far less attention has been devoted to the domain of insider trading law.3 For example, insider trading law applies to stock, but does it cover bonds?4 How about commercial real estate,5 coveted artworks,6 or copper?7 Should it? The question of domain is distinct from the questions of whether we ought to have insider trading law at all or what precise form that law ought to take.

    Most scholars have assumed a limited domain, covering just familiar securities such as common stock.8 Some scholars have argued against including other asset classes as peer markets for consideration and regulation—or deregulation—and have offered rationales for doing so.9 By contrast, I have previously argued for a wider domain, a change that has since become law,10 but I did not provide a limiting principle. In this Article, I do; providing a simple test that demarks the outer boundary of insider trading law. In building up the case for this principle, I carefully attend to assets that are commonly thought to lie beyond the domain of insider trading law and policy, and which are important in their own right: crypto assets, such as bitcoin.

    Crypto assets are new, but they are already outside the domain of insider trading law for most skeptics.11 American insider trading law regulates trading in material non-public information.12 For example, an executive might sell her stock after seeing an early draft of an earnings report. By contrast, many doubt the existence of material non-public information about open-source, virtual currencies:

    Since bitcoin is a digital asset that functions as a medium of exchange, all of the relevant information needed to price bitcoin is publicly available. For example, unlike traditional securities, there are no important periodic information events, such as earnings announcements. Since there is no “inside” information to exploit, bitcoin valuations are based on publicly available information, providing a relatively high degree of information transparency.13

    Likewise, American insider trading law generally requires the trader to have breached a duty of disclosure arising out of a relationship of trust or confidence.14 For example, a corporate executive occupies a trusted role at a corporation and, in some sense, works for the shareholders; it would be a betrayal to buy shares from the shareholders at a price they are sure to regret. By contrast, many crypto assets have no “executives” to trust or “shareholders” to betray. They are instead impersonal, decentralized,15 and “trustless.”16

    More foundationally, it is often argued that insider trading law does more harm to markets than good,17 and this might be particularly true of crypto assets. The red tape of regulation and law enforcement could constrain the experimentation and informality at the heart of this free-wheeling, open-source movement.18 Moreover, crypto assets require widespread adoption to become viable, and regulation can put a drag on such adoptions.19 Indeed, a central attraction of crypto assets for many users is that they work well even without state enforcement.20

    Crypto assets also exhibit innovative technological features that may obviate the need for familiar regulatory responses (such as securities regulation, the area most closely associated with insider trading regulation) or even render them counterproductive. For example, crypto assets are subject to a radical check on market abuse: If users dislike a set of transactions, they are free to endorse a “fork” in the chain, which would undo the disputed transaction.21 They, in effect, abandon the current asset en masse in favor of a nearly identical replacement, which differs only in that it does not recognize the disputed transaction. For example, when users of one crypto asset discovered a bug that allowed the disappearance of $60 million,22 they voted with their feet to abandon the old crypto asset and instead utilize a new asset —identical in every way except that it lacked the bug and the theft.23 This radical exit option has no analog in the world of securities, where the stock of a company retains its claims on the company’s assets even if the majority of shareholders were to dump the stock in protest over insider trading.24 Do familiar insider trading policies—fairness and the rest—make any sense when any form of market abuse is subject to radical revision? Why prioritize enforcement if the victims of insider trading can undo the offense democratically, especially when there are other real problems befalling the crypto asset market, such as market manipulation and outright fraud?25

    Thus, the existing literature makes three broad claims against the enforcement of insider trading law in crypto assets: the law doesn’t fit, it does more harm than good, and we have bigger fish to fry. Put simply, critics think that the domain of insider trading law ends far before it reaches crypto assets.

    Much of this Article is a rejoinder to that consensus: I argue that enforcement of insider trading law in crypto assets fits naturally within existing doctrine, policies, and priorities.

    Insider trading doctrine clearly applies to most familiar crypto assets and their traders. The legal requisites for insider trading regulation—jurisdiction, material non-public information, breach of duty—are frequently satisfied.26 The most obvious examples of this concern misappropriation by employees of crypto asset trading venues about the venue’s plans to support a crypto asset; allegations of this sort of insider trading have already ended up in federal court.27 But there are many more examples, such as misappropriation by government officials and members of mining pools. Ultimately the question is not whether insider trading law applies to crypto assets; it is whether we want it to.

    As a policy matter, the policies that justify insider trading law for other financial assets also apply to crypto assets: We care about fairness, price accuracy, property rights, and the rest.28 While crypto assets boast many features that could seem to make familiar regulations inapposite, in fact, these features generate new and powerful avenues for insider trading. For example, while “forking” can solve some problems, it generates others.29 Those with foreknowledge of a fork, or a market intermediary’s reaction to the fork, know about a kind of material event with no easy analogue in existing insider trading cases literature.30 Nor is regulation inappropriate in light of the innovative nature of this nascent market or anarchic values driving many crypto enthusiasts. Regulation can be useful for taming the former while respecting the latter.31

    Although much of this Article focuses on insider trading with a new asset, an examination of insider trading law and policy in crypto assets teaches us more than the right way to regulate crypto assets. Most importantly, a discussion of crypto assets gives us purchase on a general theory of the domain of insider trading law. There is a principle that links common stock and crypto assets, which are within the domain of insider trading law, but not commercial real estate and precious art and other assets, which are clearly beyond the domain. A study in the domain’s borderland helps us to be thoughtful about the lines we draw and self-conscious of the reasons for drawing them.

    Specifically, I apply a market microstructure framework to delineate the reach of insider trading law. Informed trading tends to increase price accuracy and decrease liquidity. Optimal insider trading policy is a function of those two effects: discouraging types of trading that decrease liquidity by more than they increase price accuracy. While both effects vary by type of informed trading, only liquidity effects vary greatly by asset class. Insider trading law’s domain includes those assets for which we expect informed trading to materially affect liquidity. That includes many high-volume, fungible assets such as stocks and crypto assets, but not parking lots and paintings.

    The structure of this Article is as follows. Part II provides a stylized introduction to the technology and community of crypto assets. Part III reviews insider trading law. Part IV refutes the notion that insider trading doctrine does not cover or fit crypto assets. Part V addresses some reasons that crypto assets may differ from familiar assets in terms of the policies of insider trading law, showing that these considerations can support insider trading enforcement. Part VI widens the lens from crypto assets in search of a general principle of insider trading regulation.

    Several caveats before beginning in earnest: First, this Article is not focused on many important legal and policy questions posed by crypto assets in relation to money laundering,32 custody,33 taxation,34 contract law35 and theory,36 corporate governance,37 environmental law,38 financial stability,39 law enforcement,40 national autonomy,41 bankruptcy,42 theft,43 and ordinary fraud.44

    Second, genuine data and research on crypto assets remains scarce, and the technology changes rapidly in this space,45 making it challenging to say anything both meaningful and enduring.46 This Article is meant to improve our understanding of a rapidly evolving market by shining a flashbulb at a particular shadow at a particular moment.47

    Third, this Article does not argue for a specific form of insider trading regulation for crypto assets or elsewhere. The literature on insider trading is vast and cannot be rehashed as an aside in the middle of an otherwise full paper. This Article is meant to be compatible with most debates elsewhere in the literature. When scholars call for more or less regulation of insider trading, they have in mind some domain: This Article is about defining that domain.

    II. Introduction to Crypto Assets

    Crypto assets emerge as the confluence of several important social, economic, and technological trends. New technologies for distributed and peer-to-peer networks arose just before the financial crisis of 2007 shattered public confidence in familiar financial institutions.48 A multitude of “Fintech” projects have since emerged.49 Some used technology to cut through the complexity that had confounded the banks and governments.50 Others sought to cut out the middle-man in financial markets, delivering a person-to-person experience linking borrower or entrepreneur with those who are helping to kickstart their project.51 In 2008, a white paper by an anonymous author or group of authors laid the foundations for a potentially transformative technological development of currency, property, and finance.52

    Crypto assets are a form of property distinguished by their use of a distributed ledger,53 a system by which features of the asset and its current ownership are verified and recorded semi-publicly, with no one person serving as the official record-keeper.54 Crypto assets are called “crypto” because the verification and recordation technology relies on cryptography.55 Crypto assets are sometimes called virtual currency,56 coins,57 or tokens.58 Transactions are effectuated through program functions known as smart contracts.59

    Crypto assets can be functionally organized into four overlapping types. First, a payment token is a crypto asset that is intended to be used as a form of virtual currency. For example, many merchants accept bitcoin in lieu of legal tender.60 Second, some tokens are intended to raise money for a business and entitle the user to some portion of the value of a business. Because these tokens operate as substitutes for the traditional securities (stocks and bonds) used in capital markets, these can be called security tokens. It has become common in some circles to talk about an ICO, or initial coin offering, as a public sale of coins to raise money for an enterprise.61

    Third, some tokens entitle the possessor to patronize a business as a customer or consumer. For example, Filecoin tokens entitle the user to claim a certain amount of cloud storage or cloud processing capacity from the related company, Filecoin.62 Such crypto assets are often called utility tokens. The most familiar analogs to these crypto assets are gift cards: A $25 Amazon.com gift card entitles the user to a certain amount of Amazon products in the future. Some forms of crowd-funding are also similar to crypto assets: Supporters may contribute money to a band in the hopes that they can someday hear their newly recorded album.63

    Finally, some crypto assets are defined exclusively in terms of the value of other crypto assets. For example, it is now possible to buy Bitcoin Futures on the Chicago Mercantile Exchange (“CME”). Each dollar invested delivers five times the gains or losses of owning bitcoin itself.64 This Bitcoin future is a crypto derivative asset because it derives its values from some underlying crypto asset.

    These categories are not mutually exclusive.65 A business may raise money by pre-selling gift cards for the use of its services (utility), but it may bundle with those tokens the right to a portion of the venture’s future profits (security) and the right to swap it for bitcoin at any time (derivative). Some merchants may decide to accept these tokens rather than cash (payment).

    The crypto asset drama has a cast of four main characters,66 though a given individual may play more than one role at once. Users invest in, spend, or trade crypto assets. Developers work to create, market, and improve crypto assets. Many developers are programmers who work on technical problems including code, but some developers play managerial, strategic, or communicative roles. Trading venues are web-based businesses at which crypto assets may be bought or sold.

    Finally, miners play a distinctive role in maintaining the ledger, the decentralized scorecard of who owns what. Miners are persons or corporations that own computers, which they instruct to perform computational operations essential to maintaining the ledger. Users electronically signal that they have transacted, and miners’ computers verify that the transaction genuinely took place and then record the transaction in the blockchain. For this service, they are compensated by fees, often in the form of the relevant crypto assets. The nomenclature of “mining” represents the sense in which the miners perform mundane labor, in order to realize something of value—which some other miner could reach if they “dig” more quickly or effectively in verifying and recording transactions.67

    It is often said that crypto asset transactions are irreversible and immutable,68 but this is only is half-right. The redundant records produced across multiple computers means that no single actor can unilaterally alter or conceal a record or lose it in a fire.69 However, the blockchain is not utterly immutable except insofar as miners maintain it. Which transactions are “real,” and what properties the assets possess, are matters of consensus.70 Miners decide which purported transactions to validate and which version of the ledger to validate. If miners decide en masse to negate a particular transaction or change the rules for such transactions in the future, then the crypto asset’s nature is instantly changed.71

    Any time miners adopt a new version of a preexisting chain, it is a fork.72 Forks are common means of updating the code of a crypto asset.73 If all miners implement the change, then the fork amounts to a software update.

    However, if some miners continue to process transactions under the old chain after a fork, then there are two chains. Both may persist, with independent value and a community of devoted users, or one may cannibalize attention and drive the other out of the economy. Which chain is the “real” chain and which one is irrelevant—or whether both are—is a purely social determination.74 Crypto assets therefore evolve and grow through a two-step process. Some miners vote with their mining assets to support an alternative chain, and users vote with their wallets which version of the chain to buy and use.

    Despite the word “crypto,” crypto assets are often massively transparent.75 The transaction history and current ownership of each crypto asset is available for public view.76 Although there are efforts to create assets with greater opacity,77 transparency is a crucial element in existing distributed ledger systems. If transactional details were hidden, it would be impossible for miners to conclusively decide whether putative subsequent transactions were compatible with existing endowments. For example, if John transfers all his crypto assets to Rachel on Monday and then purports to transfer them all to Nancy on Tuesday, it is essential that the latter transaction be rejected by the community.78 Only transparency makes this possible.

    These two key traits of crypto assets, permanence and transparency, interact to produce a surprisingly accountable transactional universe. “[I]f at any point your true identity is linked to a wallet address, then the whole history of your transactions then becomes public knowledge.”79 An investigator with knowledge of several people’s transactions may be able to piece together information about others’ identities.80 Thus, crypto assets are far less private than many people imagine. The impression of total anonymity is partially a product of early associations with criminal enterprise,81 and the inauspicious use of the word “crypto” in the asset’s description.

    III. Insider Trading Law

    This Part provides a brief primer on federal insider trading law. Specific prohibitions on insider trading arise under three bodies of law: securities regulation, commodities regulation, and federal wire and mail fraud. This Part reviews the various theories of liability under each body of law.

    A. Securities Regulation

    The main source of insider trading law is securities regulation, as articulated in the Securities Act of 1933,82 the Securities Exchange Act of 1934,83 subsequent SEC rules, and judicial decisions. These laws apply only to trading in securities,84 a category that includes most stocks and bonds, as well as similar assets and instruments whose value is fundamentally linked to them.85

    There are three statutory or regulatory prohibitions on insider trading in securities. Most explicitly, § 16 of the Securities Exchange Act of 1934 (“Exchange Act”) makes officers, directors, and owners of at least ten percent of a publicly-traded entity’s shares (a trio sometimes called statutory insiders)86 liable for disgorgement of any profits from trading their company’s shares within a six-month window, regardless of whether they had any kind of inside information. This “short-swing profits” rule operates mechanically to penalize rapid trading, even where it does not resemble quintessential insider trading.

    Second, Exchange Act Rule 14e-3 bars trading while in possession of material nonpublic information about a pending tender offer.87 Information “is material if there is a substantial likelihood that a reasonable shareholder would consider it important.”88 In most cases, the mere knowledge that a tender offer is forthcoming will count as material information. A tender offer is a public invitation to sell or tender securities to an acquirer,89 often in connection with an attempt to take over a company without the approval of the target company’s board.90 Rule 14e-3 applies by its terms to equity, bonds, and other securities.91

    Third, most insider trading cases are pursued under § 10(b) of the Exchange Act, a disclosure-oriented provision prohibiting fraudulent and deceptive practices.92 To pursue insider trading under § 10(b), prosecutors or plaintiffs are required to argue that insider trading is not just unfair, but actually fraudulent.93 This could be difficult since most securities trade in anonymous markets, in which traders make no affirmative representations at all, let alone false representations. Insider trading law overcomes this problem by identifying circumstances in which silence can be fraudulent. This means the law “prohibit[s] undisclosed trading on inside corporate information by individuals who are under a duty of trust and confidence that prohibits them from secretly using such information for their personal advantage.”94 Courts have developed two theories under which such a duty of trust and confidence exists, such that a silent trader perpetrates the sort of fraud banned by Rule 10b-5.

    The classical theory holds that a trader defrauds the shareholder with whom she trades by failing to disclose important information to a person for whom she is a fiduciary. The classical theory primarily contemplates inside trading by an officer or director,95 who can be said to indirectly work for and manage property on behalf of her shareholders.96 The insider occupies a sort of trustee role and should not take advantage of her beneficiaries by trading ruthlessly with them, using knowledge she only has because of her entrustment.97 The classical theory also covers corporate advisors,98 investment advisors and other financial professionals,99 and government officials who owe a duty to all citizens not to misuse their offices for personal gain.100

    The misappropriation theory holds that a trader who feigns loyalty to a company or person to gain access to secrets ultimately defrauds his source out of information when he misuses the information for trading.101 It bars trading whenever someone learns information in a context that implies confidentiality, even if the trader is not a corporate insider. For example, the misappropriation theory is violated if a member of Alcoholics Anonymous trades based on information learned at their confidential meetings,102 or a broker front-runs (i.e., places trades ahead of) their client.103

    B. Commodities

    Commodities and futures are subject to the Commodity Exchange Act of 1936 (“‘36 Act”) and the jurisdiction of the Commodity Futures Trading Commission (“CFTC”). Commodities are broadly (and somewhat circularly) defined as “all services, rights, and interests . . . in which contracts for future delivery are presently or in the future dealt in.”104 This covers familiar commodities such as copper and corn, but also abstract and financial material such as interest rates,105 foreign currencies,106 and baskets of stock.107

    While insider trading in commodities was permitted under the ‘36 Act for all of the twentieth century,108 the CFTC adopted rules in 2011 that seemed to import much of the insider trading jurisprudence from securities regulation.109 Since that time, the CFTC has brought two insider trading cases.110 The post-2011 insider trading regime in commodities is commonly said to track the misappropriation theory from the regulation of securities,111 though some aspects of the regime identify duties of particular individuals —brokers,112 exchange officials,113 government officials114—as subject to a status-based duty, akin to the classical theory.

    C. Mail and Wire Fraud

    The Department of Justice can bring insider trading cases under the federal mail fraud115 and wire fraud statutes.116 The elements of the charge are substantially similar to those of the misappropriation theory under SEC Rule 10b-5.117 However, mail and wire fraud jurisprudence covers cases that 10b-5 does not.118 The most important difference for the purposes of this Article is that federal prosecutors can bring cases that do not involve trading in securities or commodities.119

    IV. Doctrine: Do Crypto Assets Fit Insider Trading Law?

    It is common to believe that insider trading law and crypto assets do not fit together.120 The main insider trading theories for securities require the breach of a duty of trust or confidence and material non-public information, but many crypto assets seem to lack “issuers” or “shareholders” whose trust can be betrayed, or officers or directors who can commit a betrayal.121

    Other skepticism arises out of issues that are distinctive to crypto assets. Some wonder whether crypto assets fit into any regulatory box subject to insider trading law.122 Crypto assets are supported by software that is open source, so what “non-public” information is there? And just what is material to an asset as speculative as Bitcoin or as fanciful as some of its lesser known competitors, such as CryptoKitty?123

    This Part shows that the law of insider trading can, and in many cases does, apply to cryptocurrency. Although the three key issues (jurisdiction; material non-public information; and duty) are treated separately below, it is worth keeping in mind the following case in which all three allegedly came together. In November 2017, a smallish cryptocurrency’s price soared on good news—an important trading website (Coinbase) would soon support it. Before making the announcement, Coinbase’s executives bought vast sums of the favored cryptocurrency (Bitcoin Cash). Trading in advance of an announcement violated company policy.124 One trader sued alleging insider trading.125 If true, the Coinbase incident would satisfy the elements of a familiar misappropriation-theory insider trading case—as the discussion below demonstrates.

    This Part does not argue that the law should apply in any given case or any cases at all. That policy discussion exists in Parts V and VI. Rather, the point is that cryptocurrency is a perfectly sensible subject of insider trading regulation, and it is a policy decision whether to ratify that existing status.

    A. Regulated Subject Matter

    Some have questioned whether insider trading law even applies to crypto assets, since the focus of American insider trading jurisprudence has concerned common stock in publicly traded companies, while crypto assets are something else entirely. These arguments are plainly wrong—it is obvious that crypto assets are subject to at least enough of the insider trading jurisprudence to allow federal prosecutors to bring successful criminal actions.126

    First, federal mail and wire fraud statutes apply to crypto assets. That is because federal mail and wire fraud statutes apply to insider trading in any asset, be it a security, a commodity, or a fanciful crypto asset. The U.S. Supreme Court in United States v. O’Hagan affirmed a criminal conviction of a lawyer who misappropriated material non-public information from his law firm and its clients and used it to trade.127 In that case, the traded subject matter was stock, but there is nothing in the jurisprudence that limits the application of mail and wire fraud statutes to securities—and non-insider trading cases have used those statutes in numerous non-securities contexts.128 Much of federal insider trading law applies regardless of the type of asset at issue, and crypto assets are certainly assets.

    However, it is also worth examining why many crypto assets are subject to securities and commodities regulation (or both) with their attendant insider trading rules. This is because characterization of crypto assets as a security or commodity would empower civil enforcement by the SEC, CFTC, and private plaintiffs. It also unlocks additional grounds for liability.

    SEC Chairman Clayton made waves in November 2018 when he seemed to announce that Bitcoin is not a security.129 This is significant because Chairman Clayton was for characterizing crypto assets as securities before he was against it.130 Despite the Chairman’s statement, the SEC has taken numerous other steps to clarify its position. In 2017, a SEC report concluded one prominent set of crypto assets, DAO Tokens, were securities.131 Further, the SEC has asserted that crypto assets are securities in enforcement actions against promoters of crypto assets132 and actions against trading venues for crypto assets.133 Most analyses—by the SEC and others—have highlighted the substantial likelihood that any given crypto asset is indeed a security subject to securities regulation.134

    To the degree that analysts conclude that securities laws are inapplicable, it tends to be regarding crypto assets that function more purely as a currency.135 Such a conclusion, however, takes crypto assets out of the frying pan of securities regulation and into the fire of commodities regulation, which includes a similar set of insider trading rules.

    Given the expansive definition of “commodity” and given the similarity between many crypto assets and currencies (which are subject to the CFTC’s jurisdiction), it is highly likely that any given crypto currency is subject to the anti-fraud provisions of the Commodity Exchange Act.136 The CFTC has asserted,137 and two federal courts have held,138 that virtual currencies such as bitcoin are commodities for the purpose of the Commodity Exchange Act.139 The CFTC can prosecute fraud in both the spot market (the crypto assets themselves) and derivative contracts arising from them.140

    Recently, there have been legislative efforts to exclude some crypto assets from the coverage of the securities acts.141 Whatever the prospects for these bills, their protections are not retroactive.142 More importantly, there has been no effort to remove crypto assets from the ambit of the Commodity Exchange Act or wire fraud and mail fraud. Insofar as the Commodity Exchange Act also regulates insider trading and also applies to crypto assets, recent legislative fixes do not extinguish the need for insider trading analysis.

    B. Material, Non-Public Information

    The touchstone for insider trading regulation in any form is the existence of material, non-public information.143 Some examples of this are familiar to securities lawyers, such as a corporation’s quarterly earnings report. But there is material non-public information even for crypto assets that do not neatly analogize to securities. The SEC addressed the issue of material non-public information in rejecting a proposal to let several securities exchanges begin trading shares in the Winklevoss twins’ Bitcoin Exchange Traded Fund (“ETF”):144

    Assuming there is no inside information related to the earnings or revenue of bitcoin, there may be material nonpublic information related to: the actions of regulators with respect to bitcoin; order flow, such as plans of market participants to significantly increase or decrease their holdings in bitcoin; new sources of demand, such as new ETPs that would hold bitcoin; or the decision of a bitcoin-based ETP, a bitcoin trading venue, or a bitcoin wallet service provider with respect to how it would respond to a “fork” in the blockchain, which would create two different, non-interchangeable types of bitcoin.145

    Some of these additional considerations noted by the SEC are also familiar to securities lawyers: news coverage, regulatory treatment, exchange treatment, and trade data. All four of these familiar forms exist for crypto assets and ordinary assets. All these forms of material information are discussed below. In addition, the SEC notes arguably novel forms of material non-public information relating to forks. These are discussed both here and partially in Part VI.146

    1. Issuer Information

    For security tokens, which are functionally similar to securities, a whole ambit of information about the issuing company is plainly material and non-public.147 The financial condition of the company issuing tokens is clearly material to the buyer of a token. So are the company’s business prospects and legal risks. Securities lawyers spend their careers opining on the many forms of information generated by a company that are material to its investors. Given that almost no issuers of tokens are in the habit of periodic disclosure, such company information is usually non-public.

    Two practitioners recently noted another item of material non-public information that may apply to many crypto assets but has been neglected in many cases: lockup agreements, or restrictions on resale. This information is material because the expiration of a lockup often coincides with a substantial increase in marketable assets, putting downward pressure on the price.148 When companies pay for acquisitions with securities,149 the lockups imposed on those securities is obviously material to investors.150 It even gets its own line in many familiar disclosure documents.151 When companies waive their lockup period for someone, they promptly announce it on an 8-K.152

    Lockups are common in crypto assets: “About . . . 15 percent of projects had a lockup period of 1 to 3 months, 17 percent locked their tokens up for 4 to 6 months, 14 percent had a lockup period of 8 to 12 months, and 6 percent of projects studied locked their tokens up for 18 to 24 months.”153 They may lock up a substantial portion of the supply of crypto assets.154 Yet organizations that use tokens to acquire assets and companies do not seem to disclose any lockups on resale of the acquisition consideration, and may forbid investors from disclosing the nature of lockups.155

    Companies and individuals who trade during the lockup period do so while in possession of material non-public information, even if they are not themselves subject to the lockup. For example, suppose a venture capitalist buys crypto assets knowing that the founders are subject to a nine-month lockup. The venture capitalist sells her crypto assets eight months later, shortly before the founders become eligible to sell. The venture capitalist has traded while in possession of material non-public information and could potentially be liable for damages in a private securities suit to any contemporaneous trader, or in a government enforcement action.156 Likewise, insiders who learn that a lockup period is going to be modified have material non-public information until that modification is disclosed.157

    2. Media Coverage and Commercial Treatment

    Positive and negative news coverage can affect the price of an asset and is plainly material for the purposes of insider trading law.158 For example, in United States v. Carpenter,159 the Second Circuit upheld the conviction of one of the authors of the Wall Street Journal’s “Heard on the Street” column on an insider trading offense.160 The majority rejected the dissent’s notion that information is only material if it is “securities-related.”161 The court instead held that any information is material if it “in reasonable and objective contemplation might affect the value of the corporation stock or securities.”162

    Media coverage of crypto assets frequently impacts their price.163 News that blue-chip companies like Subway and Microsoft would accept bitcoin pushed up the price of bitcoin.164 Bitcoin fell five percent when Business Insider reported that Goldman Sachs would abandon plans to open a cryptocurrency trading desk.165 Given this responsiveness to media coverage and breaking news, one with foreknowledge of such news would be able to profit predictably.166

    3. Regulation and Enforcement

    Actions by regulators also have the potential to affect the price of crypto assets. When regulators authorize bitcoin as a lawful payment method, the price goes up.167 Regulation with the opposite force pushes the price back down.168

    Earlier, we quoted the SEC’s decision to reject the Winklevoss ETF.169 That decision both addressed material non-public information and constituted it: The price of bitcoin fell three percent immediately after the decision was announced. Indeed, this rebuke was part of a series of rebukes, with at least one leading to rumors of insider trading. On August 4, 2018, bitcoin prices dropped while ether, another cryptocurrency which often moves in tandem, stayed still. Three days later, the SEC announced that it would need more time to evaluate a proposed bitcoin ETF listing on an exchange.170 Many Bitcoin enthusiasts inferred that the SEC’s delay meant bad news for the eventual approval of the product,171 which had been previously rejected in its bid for New York Stock Exchange listing.172 It was later rumored that the price drop came from traders who somehow learned of the SEC’s pending decision.173 In each of these cases, government employees or their tippees would have had foreknowledge of the action.

    However, even if government employees do not trade, those close to a company may have foreknowledge of regulation or enforcement due to their interactions with the government. For example, regulatory action often follows problems or scandals at bitcoin platforms.174 Insiders at the scandal-ridden company would have been in a position to trade, knowing that new regulation or enforcement action was impending.175

    4. Exchange Listings

    Listing a crypto asset for trading on an exchange or trading venue can have a big effect on the price of that crypto asset.176 Trading venues and their insiders have “access to non-public news (like the impending listing of a new virtual currency on the platform.”177 When Coinbase announced it would support Ethereum Classic, the crypto asset’s price rose 20 percent.178 In another notorious incident, cited at the beginning of this Part, Coinbase announced on twitter that it would support trading in Bitcoin Cash—after having repeatedly denied that such support was forthcoming.179 Just before the announcement, the price of Bitcoin Cash surged to an all-time high, consistent with someone at Coinbase trading on or leaking the information.180

    For all that rides on a listing, trading venues do not all observe best practices for making and disclosing their decisions, as Coinbase’s tweet-based change implies. Trading venues’ procedures are often opaque and discretionary.181 New York’s Office of the Attorney General summarized things this way:

    Across the board, the OAG found that platforms’ determinations of whether to list a given virtual asset were largely subjective. No platform articulated a consistent methodology used to determine whether and why it would list a given virtual asset. Some objective factors did appear to be considered by many. For instance, platforms often look at the total value or “market capitalization” of a virtual asset, or its average daily trading volume. But the OAG found there is no rhyme or reason to how those objective factors are applied, and there is certainly no consistent application across platforms.182

    Nor is the uncertainty located just at the bottom of the pile. The second largest crypto asset by market capitalization, at the time of writing, remains untradeable at Coinbase.183

    Similar uncertainty and opportunity surrounding listing decisions applies to platforms listing derivatives on crypto assets. In many cases, exchanges provide little ex ante guidance about whether they will list or delist a crypto asset,184 even though these decisions can influence the price of the underlying crypto asset.185 LedgerX, the first clearinghouse and derivatives exchange authorized by the CFTC,186 has an official policy on its treatment of new assets produced by forks: “Our management and risk committees will evaluate each hard fork on a case by case basis. We will then publish public notices to our members as to our plan for that hard fork as soon as prudently possible.”187 Although three factors are listed as relevant to this evaluation (marketplace support, feasibility and security, and regulatory comfort), they are hardly algorithmic.188 LedgerX retains substantial discretion in whether to list derivatives, and LedgerX insiders will know before the market which way that discretionary process is leaning. Also worth noting is that LedgerX promises to notify its “members” as soon as prudently possible. But selective disclosure to members gives them an edge over non-members in trading crypto assets whose value may depend in large part on their treatment by LedgerX.189 Long before the derivatives contract is added or removed, LedgerX members get a chance to opportunistically trade with non-members in the spot market.

    LedgerX’s policy is largely discretionary, but more objective procedures do not eliminate the possibility for informed trading—they simply change how it happens.190 A discretionary standard gives insiders at LedgerX (and, later, members) a leg up. A mechanical standard gives an advantage to whomever has early access to the data utilized in that mechanical standard.191

    For example, the CME Group (formerly CBOT)—the oldest and perhaps most important derivatives exchange192—has publicized its criteria for adding new crypto assets for derivative contracts.193 One condition is that the new asset must be traded on at least two of the spot markets recognized by the CME. That means that insiders at the spot market have foreknowledge about whether the asset will be listed by the CME. Once the second exchange agrees to do so, it may be certain that a powerful form of support for the asset will become available.194

    Of course, the decision to list a derivative is only one discretionary choice by a derivatives exchange. They may also update their calculation methodology. For example, introduction of a new price feed into the benchmark can greatly and predictably change the settlement price.195 They may also decide whether to discontinue a product196 or even halt trading.197 A trading halt is inconvenient for traders in an asset, may limit the expression of pessimistic views, and may signal the exchange’s pessimism about the viability of the asset.

    5. Trading

    Information about planned trades can be material because a large purchase or sale can move market prices. Brokers are often tempted to “front-run” their clients by placing orders likely to pay off in light of the following order,198 and clients themselves sometimes plan to “self-front-run” by anticipating the effect of their own orders on the market.199 The same dynamics are likely feasible in crypto asset markets,200 with an additional possibility. When users transact crypto assets, miners record the transaction in the blockchain. There exists a window of time in which a miner knows about the transaction prior to recordation. A miner can decide in that moment whether to initiate their own transaction and insert it into the block prior to the temporarily prior one. They can use this to make a trade in light of information coming to market, or to literally usurp the very transaction they were meant to record.201

    Even without front-running, there are plenty of other ways to gain from information about trading plans. Trading venues know something about the otherwise undisclosed identity of traders.202 For example, in September of 2018, the co-founder of the third largest Crypto asset, Ripple Labs’s XRP, began selling vast sums of his personal stake in the asset.203 XRP declined 13 percent following the news.204 The founder had announced similar sell off plans in 2014, resulting in an even bigger drop.205 A broker or platform may know before others that large sales are coming from a founder and trade on the impending negative market reaction.

    Even completely anonymous, aggregate trading data can be valuable. Aggregated order flow data helps traders in ordinary currency to outperform the market.206 Trading venues now show great interest harvesting and selling equivalent data for crypto assets.207

    Another crucial form of information known to markets is when orders are timed. For example, it is common for traders to place orders to execute only at a certain time of day, usually at closing or the moment at which a benchmark is set.208 Net order flow at that moment can influence the benchmark price, and the value of any derivative contract based on it.209 The same is true for crypto assets. The settlement value for Bitcoin futures is based on the average of the price of five spot markets, derived at a designated moment. Those five spot markets may be able to observe pre-set orders well in advance of the fixing moment. They can make an educated guess on how one of only five data points will resolve, and therefore what futures are worth.

    * * *

    There are plainly many forms of material non-public information bearing on the price of crypto assets. Indeed, scholars have already taken steps to quantify the price impact of material non-public information on crypto assets. One paper identified dozens of incidents in which the price of a crypto asset moved substantially, seemingly because of disclosure of news, which was known privately prior to the disclosure and where there appears to have been substantial pre-disclosure trading in precisely the ways one would expect if some traders had foreknowledge of the news.210 “[T]he average estimated profit of Bitcoin informed trading is between 100,922 and 915,455 USD per event; and between 222,973 and 2,367,409 USD per large event.”211 Many of the types of information noted in this Article fall into the categories discussed above. This leaves the question of whether it is legal to trade on the forgoing information, which is often a matter of duty, discussed below.

    C. Breach of Duty

    For the most part, American law bars only trading in breach of a duty of trust or confidence: “[T]he fiduciary concept is based fully on trust—one party entrusting another to make decisions on her behalf.”212 By contrast, it is common to refer to cryptocurrency systems as “trustless.” This may suggest that there is no duty to support insider trading regulation. That said, relationships of trust and confidence are widespread in the crypto asset economy. The following shows many examples of duties of trust and confidence—or facts that permit liability without such a showing.

    1. Classical

    Some crypto assets are issued as equity securities with officers or directors. For such crypto assets, the classical theory applies as is conventional: The officers and directors of the issuer owe a duty to the shareholder-traders of the crypto assets as a result of their common relationship to the issuing firm.213 In other cases, where the crypto asset network is radically decentralized and there is no one analogous to an officer or director, there remain at least two ways the classical theory may still apply.

    First, government actors have a classical duty to abstain from certain forms of trading under the STOCK Act of 2012. This law provides that all members and employees of Congress and all other federal officials and employees owe a “duty of trust and confidence” to the “United States Government” and to the “citizens of the United States” respecting “material nonpublic information” derived from their official positions or gained through performance of their official responsibilities.214 Insofar as they learn about upcoming regulatory changes and enforcement actions, or obtain information from entities subject to regulation and investigation, it is illegal for them to trade securities and commodities (crypto or otherwise) on that basis.215

    Second, brokers and exchanges have long been held to a classical-like duty to forswear trading on customer-specific data or changes in the exchange’s policy. The same could plausibly hold for intermediaries in crypto asset trades.216

    A third rationale is that developers may owe a classical theory duty to the holders of the crypto assets they develop. Some scholars have prominently argued for this duty.217 Such a view would make it a violation of duty for a software developer of virtual currency technology to exploit any information which is material and non-public. Arguably, bugs in the code would qualify for such characterization. For example, in a highly publicized event, the DAO (“Distributed Autonomous Organization”) was hacked and about $55 million stolen.218 A flaw in the code permitted this to happen. A coder who participated in the construction of this flawed code might be liable for negligently causing the flaw (Professor Walch’s point).219 But that same public obligation to users to ensure good code may also bar trading on bad code that nevertheless goes unrepaired.

    As greater attention comes to the crypto asset sector, we discover more cases of developers learning about and intentionally concealing troubling problems with code until repairs are completed.220 These decisions may be well-intentioned, but developers who buy or sell crypto assets in the meantime may potentially run afoul insider trading norms.221

    2. Misappropriation

    As already discussed, Coinbase’s decision to list Bitcoin Cash on its platform had a large impact on the price of that crypto asset and others. Rumors swirled in the cryptocurrency community of the possible trading by Coinbase insiders, leading to a public statement by Coinbase’s CEO announcing a company policy barring trading on material non-public information.222 These platforms differ greatly in the restrictions on employees.223 At any platform with a trading policy like Coinbase’s, violation of that policy would support a finding of liability under this misappropriation theory.

    The misappropriation theory is not limited to just misuse of information by exchange employees. Any time someone has material non-public information, confidants who misappropriate it are potentially liable for insider trading. That certainly means agents of a trading platform (officers, directors, employees). It likewise means that the agents of large traders are liable when they trade on the basis of private trading plans or proprietary research.224 So too are journalists who trade on the basis of a story,225 regulators who tip friends in advance of an enforcement action,226 or employees of a company that plans to announce greater or lesser support for a crypto asset.227 There are as many ways to misappropriate information as there are to generate it.

    It is common to think that crypto assets without an “issuer” do not have a source to whom an agent might owe a duty. The forgoing examples already dispel such a thought, but distinctive features of crypto assets actually indicate that there are more sources (whose defrauding can establish a misappropriation claim) with crypto assets than traditional assets. With common stock, there is just one company that has privileged information about its own plans. Will it dilute its own equity? Will it announce low earnings? Insiders at the company can defraud the company of this information, but no others will typically have this information.

    This is not so for crypto assets, where several miners collectively perform the operations necessary to preserve and update the blockchain. Less than a dozen mining pools control 80 percent or more of the computing power that governs any given crypto asset.228 Each one of those mining pools knows much better than the market whether it will support any proposed changes to the crypto asset, and it knows slightly before others what transactions have been executed. Each pool is therefore a principal with material non-public information. If agents at any pool use this information without permission to trade, they are culpable under the misappropriation theory.

    There are still other ways for traders to misappropriate information. Several platforms deny that they engage in proprietary trading on their own account.229 If they nevertheless trade, they could be liable for fraud. Insofar as the assurance of non-trading led customers to share their data with the platform, their information would have been misappropriated by the platform.230 Platforms have foreknowledge of the trade requests of their customers. Trading ahead of such requests could easily defy the explicit or implicit assurances of the platform. Front-running is a form of market abuse that is partially coextensive with insider trading, and it has already been alleged in one platform.231

    3. Tender Offer Trades

    Not all insider trading theories require material non-public information or a breach of duty, but the few commentators to remember this in the context of crypto assets have quickly dismissed their importance.232 This is a mistake. Two insider trading theories other than the familiar duty-based theories also apply to crypto assets.

    Rule 14e-3 bars trading even on authorized information about an undisclosed tender offer,233 and it applies to any security.234 Insofar as tokens are securities, it applies to them. A tender offer for part of a security crypto asset would be subject to these rules.

    It may seem fanciful to contemplate tender offers for crypto assets,235 but such strategies are already in practitioner toolkits. For example, tender offers might be used to call in non-compliant tokens issued in the wild days of 2017 and 2018, in return for properly registered tokens:

    To start, issuers of unregistered security tokens (let’s call them “old tokens”) would have to complete a formal SEC registration process for what are essentially replacement tokens (“new tokens”). Upon the approval of such a registration, issuers would have to swap old tokens for new tokens for all willing takers—a digital tender offer of sorts.236

    A fine plan, perhaps, but it creates insider trading liability for anyone who trades on the eve of such a tender offer—including friends and advisors to the offeror who have been authorized to trade and those whose trades have nothing to do with the tender offer.

    Insider trading regulation may also apply to impede efforts to construct investment funds in crypto assets.237 Owning the underlying crypto assets creates security risks and payment challenges. Many investors will find it more familiar and convenient to buy shares in an investment fund that own crypto assets.238 However, the SEC has shown reluctance to allow such products.239 There is a workaround to sidestep the SEC’s reluctance, however: a “tender offer closed end funds.”240

    A mutual fund is a regulated investment vehicle that permits its investors a right to redeem their shares at the end of the day for their pro rata share of the net asset value of the fund.241 A closed end mutual fund is a mutual fund that denies its investors the right to make daily redemptions.242 Closed end funds can go for decades without redeeming shares or paying dividends to their investors. However, one type of closed end fund (a tender offer closed end fund) seeks to enjoy the benefits of being closed while still providing reasonable options for their investors to recover their cash when needed. They do this by periodically offering to repurchase shares from their investors pursuant to a tender offer.243 For example, the board of a closed end fund might offer every six weeks to redeem up to 25 percent of any investor’s shares.244 A fund that purchases crypto assets might be able to operate as a closed end fund with liquidity through tender offers, and thereby sidestep many regulatory hurdles.

    Tender offers can therefore be used to support the existence of ETPs for crypto assets but doing so implicates the restrictions of SEC Rule 14e-3. Anyone who has knowledge of a closed-end fund’s plans to provide liquidity by way of a tender offer cannot trade on the basis of that information, even with the fund manager’s permission.245

    4. Short Swing Trades

    Equity securities are subject to § 16 of the Securities Act of 1933, which penalizes rapid trading in equity securities by certain statutory insiders, including those who own ten percent of the stock. Essentially, it requires that profits made within a six-month window be disgorged. Section 16(a) requires prompt disclosure of any trades. There is no requirement that the trader know material non-public information, nor is there any argument that the source of information can relieve the trader of the consequences of the trading.

    Large traders who come to own ten percent of a class of crypto assets (assuming that class qualifies as an equity security) would therefore be required to file documents with the SEC documenting every single trade they make.246 They would also need to disgorge any profits made within a six-month window. To my knowledge, no trader has ever documented a trade in crypto assets with the SEC.

    This problem is likely to grow as time goes on. At present, the best known crypto assets are “proof of work” systems that reward miners who devote computing power to maintaining the system.247 But the Ethereum Foundation hopes to transition ether (arguably the second most important crypto asset after Bitcoin) to a “proof of stake” system in the near future.248 A proof of stake system rewards traders who accurately verify transactions and have bet a large amount of crypto assets that they are correct. To play the mining game in that brave new world will require substantial ownership. However, whomever buys ten percent or more of a proof of stake equity token will be essentially precluded from trading it. The implications of existing insider trading law may therefore grow considerably in the coming days.249 Those who own enough to mine may be precluded from timely selling the rewards of their labor.

    * * *

    The point of the forgoing analysis is not to argue that any particular instance of insider trading occurred or is subject to liability under the law. Nor is the point that the law ought to operate the way that it does. So far, the point is only to dispel the sense that the old categories somehow don’t apply. There is a long history of dismissing the viability of insider trading categories to currencies and commodities.250 The point of the foregoing analysis is that we cannot proceed so blithely. We must decide whether insider trading law ought to apply the same way to cryptocurrency as other assets, even as we decide what our insider trading law ought to be. It is to that we now turn.

    V. Policies & Priorities: Do Crypto Assets Need Insider Trading Law?

    The desirability of insider trading in securities law has been hotly contested for decades. Advocates for deregulation assert that insider trading improves price accuracy and managerial incentives.251 Advocates against deregulation dispute these claims,252 and argue that insider trading is unfair to ordinary investors,253 raises trading costs,254 and constitutes a theft of property from the corporate issuer.255

    Whatever the proper resolution of that debate, the crypto asset market is not somehow exempt from consideration. To the contrary, most of the policy rationales for and against insider trading law in securities and commodities apply to crypto assets as well. Indeed, some apply even more strongly. This Part presents familiar policy arguments relevant to the regulation of insider trading (such as fairness, price accuracy, and trading costs) in answer to three arguments often raised in opposition to insider trading law to the crypto asset market.

    First, it is commonplace to argue that crypto assets are in a nascent stage and that their growth and innovation requires lawyers to keep their hands off.256 A closely related point is that regulation can undermine efforts to establish a minimum effective scale of use, which crypto assets require for viability.257 Against this, I argue that regulation and enforcement can encourage growth and adoption by removing the risk and cost associated with market abuse.

    Second, while the availability of radical self-help through forking may seem to empower crypto asset users to solve their own market abuse problems without state assistance, I argue that forking creates entirely new problems, including new dimensions of insider trading.

    Third, users of crypto assets may be ideologically opposed to legalistic interventions. They prefer to live and die by their code, and insider trading may be one of many risks they’d rather handle on their own. Or so the argument goes. Many actual and would-be crypto asset users do not feel this way, and the law should consider their expectations too. In any case, I argue that even “crypto-Crusoes” benefit from a legal system that treats their values as important—insider trading law can do that.

    Fourth, crypto assets are currently plagued by fraud and market manipulation, but action against insider trading may well reduce those other ills.

    A. Red Tape, Liquidity and the Hinman Paradox

    It is now common to think that disruptive businesses grow best when they ignore laws.258 Brilliant innovators feel the need to “Move Fast and Break Things.”259 They must not let legal niceties crowd out technological and economic insights. These arguments take on greater resonance because of the Hinman Paradox, identified by Professor James Park.260

    The paradox takes its name from William Hinman, the SEC’s Director of the Corporate Finance division, who gave a speech explaining why the SEC polices unregistered sales of nascent crypto assets even as it tolerates sales of the largest unregistered crypto assets, such as bitcoin. Hinman set out a standard that implied that the SEC’s job is done once a crypto asset becomes widely used enough that it no longer relies on a few promoters for success: “If the network on which the token or coin is to function is sufficiently decentralized—where purchasers would no longer reasonably expect a person or group to carry out essential managerial or entrepreneurial efforts —the assets may not represent an investment contract.”261

    Hinman’s standard creates a chicken-and-egg problem that pits law against viability. As Professor Park puts it, “for a utility token to be distributed freely without regulation by the securities laws, it must be functional. But many utility tokens are only functional if they are distributed widely enough so that a de-centralized system arises.”262 The reason that they are only functional if they are widely distributed is that,

    “for many token platforms to operate efficiently, the platform must algorithmically or otherwise generate and pay tokens to miners, oracles, verifiers, or others who provide valuable services to the platform and the broader token ecosystem.” Therefore, “these tokens must be capable of being delivered to any person . . . and must be freely tradeable upon receipt. . . .”263

    Functional platforms reward miners with coins and these coins are only suitable inducement if salable. They are especially salable if they do not require registration, but they require registration if they are not yet functional. The Hinman Paradox is that assets escape regulation through widespread use, but they cannot achieve widespread use if they are regulated.264

    While Park is right to notice the Hinman Paradox, the analysis is incomplete because it presumes regulation is an impediment to the liquidity needed to make a system functional. In fact, market regulation is supposed to improve liquidity.265 Investors are more likely to buy if they do not fear information asymmetries or if trading costs are low. Market regulation generally advances these goals.266 In particular, insider trading law enforcement lowers trading costs,267 and concern over trading costs is now one of the most promising theories for justifying insider trading regulation.268

    Moreover, many crypto enthusiasts don’t just desire viable assets—they want viable money.269 For non-state money, the urgency of improving liquidity is even higher. As Professor Gorton and others have shown in their analysis of the rise and fall of mortgage-backed securities as a form of private money, the crucial factor in the success of the instrument is its information insensitivity.270 Informationally insensitive assets are ones for which it is very difficult to have any kind of informational advantage about its quality. All mortgage-backed securities (“MBS”) look pretty much alike—universally and terrifically complicated—so market participants mostly treated them as fungible. If they were simpler assets, payees might have been apprehensive before taking MBS as payment or collateral; is the seller only offering this MBS to me because she knows that it is about to default on its payment obligations? For an asset to work as a liquid unit of payment, recipients must not think there is any point in researching the asset’s quality,271 or any fear that the seller is foisting a lemon onto them.272 If they did, then every transaction would include a friction of due-diligence. Put simply, people must trust that their money is valuable without having to check. For money to work, a combination of economic and legal factors must reduce the gains of research to less than the cost of research. Gorton shows how this combination was achieved with asset-backed securities.273

    The stakes are equally high for crypto assets that want to be spendable.274 If the goal is to be realized, economics and law must work together to eliminate the recipients’ fear that the spender knows something bad about the asset. Insider trading law is calibrated to remove that worrisome information asymmetry. A legal process used to make assets more informationally insensitive and support their money-like attributes.

    Thus, regulation isn’t an impediment to widespread distribution of crypto assets; it is an important bridge to making that transition. Given that, the Hinman Paradox (regulated unless functional, functional only if widely used, regulation impairs use) must be supplemented by what we can call the Hinman Corollary: Regulation can help a token to be widely used. The Hinman Paradox poses a puzzle: How do you escape the pull of regulation if the only way out is through? The Hinman Corollary reconstructs the puzzle: Having achieved sufficient scale so as to be functional and thus avoid regulation, how do you not lose the regulation that preserves that scale? This question looms large under Hinman’s approach, but it is quite the opposite problem initially posited, and it is no argument against law enforcement.

    B. Forks

    It is often asserted that regulation is less necessary for crypto assets because any problematic transactions can always be erased by the consensus of the community. For example, the infamous DAO token hack involved the extraction of almost $60 million of ether from an investment community,275 but it was effectively erased when miners and users forked the code to a new version of ether that was the same except in that the attack did not take place.276 The crypto-burglars still had the ether they stole, but in a form the community ceased to recognize. It is the equivalent of a community boycotting all the dollars stolen from a bank—who needs cops if this amazing form of self-help is available?

    While the self-help possibilities for crypto assets are potentially transformative, there are four reasons it would be premature to end familiar forms of law enforcement at this time. First, the decision to build community consensus around a fork is costly.277 The decision to alter block size for bitcoin (which created Bitcoin Cash) took years of negotiations and troubleshooting, including numerous international meetings for developers and miners.278 These decisions are contentious,279 and it is inefficient to incur such decision costs every time there is alleged wrongdoing. There is a reason that we do not ask for a national referendum in every criminal trial—it pays to delegate to a professional law enforcement system.

    Second, forking poses distinctive costs to users of the crypto asset. In the days leading up to the potential fork, uncertainty reigns as users are unsure whether their asset will be changed or useless as a result of the fork. After the fork, it is common for competing versions of the (similarly named) asset to trade simultaneously, depressing the price of both. Forks that annul fraudulent transactions may have innocent victims, such as those who received payments that are no longer recognized by the community.

    Third, and most interestingly, radical self-help in the form of forking unlocks a powerful new form of informed trading, without a perfect analog in the securities realm: Miners and users know whether forks will occur and whether they will succeed before other users do, because they help make the decision.

    To understand this point, begin with a consideration of voting. Individuals always know before others how they will vote. In corporate elections, a large shareholder’s voting plans may be material non-public information, because it can foretell the future of a corporation.280 Will bad management be retained? Will a proposed merger collapse? These considerations have a big impact on share price, and a large investor can decide a close vote. That large investor necessarily knows better than third-parties how its vote will be cast. There is an ineradicable form of asymmetric information where a few people make important decisions.

    Crypto assets also involve votes of a kind, and thus empower powerful voters to profit from their foreknowledge of the vote’s likely result. Crypto assets do not strictly operate on democratic principles, but a parallel system applies in the form of miners’ “votes” on which blockchain to validate and users’ “votes” on which crypto assets to value and patronize.281 In both cases, prior knowledge of how patrons of the asset will respond to a conflict can prove important in predicting the future price of assets.282

    Consider, for example, the Ethereum DAO fork, in which users decided whether to retain their existing support for Ethereum or switch their efforts over to a new version that erased the harmful effects of a bug-related theft. “The preparations for the hard fork included . . . an advance poll of the Ethereum miners to see how likely the hard fork was to succeed. Only a very small percentage of ether holders or miners voted in the advance polls, but the Ethereum developers decided to proceed with the hard fork.”283 Presumably a large miner would not need a poll to have a decent guess at how she would vote and how the vote itself might come out.

    Those guesses might be highly accurate, given that mining is a highly concentrated operation.284 As of summer 2018, two subsidiaries of a single corporation controlled 42 percent of all Bitcoin mining capacity.285 Another 40 percent is controlled by just four more firms.286 Concentration is even greater at other crypto assets: Only two miners make up the majority of Ethereum mining.287 The rapid cartelization of the mining field means that just three or four teams are in a position to unilaterally control the future of many crypto assets, and the cooperation of a dozen or so would mean that any resulting fork would see no substantial opposition.288

    It is possible to profit based on predictions about how miners will vote even when decisions concern crypto assets that do not yet exist, since derivatives contracts often trade long before the fork occurs. For example, in the recent fork in Bitcoin Cash between large and small block size advocates289 wild price swings in the futures market ahead of time were driven largely by expectations about which powerful miners and platforms supported which form of the crypto asset.290 Insiders at such a mining operation or platform could guess better than others where their support would ultimately fall, and thus the true price of the asset underlying the futures contract.

    C. Crypto Anarchism

    Some crypto enthusiasts are driven by their desire for, or belief in the inevitability of, the demise of the familiar banking system,291 state-issued money,292 or of states altogether.293 Others value secrecy far more than price and convenience.294 They are disproportionately libertarian.295 So it may seem ironic to bring the power of the state to defend this community from insider trading.

    Relatedly, many crypto assets have been developed as open source projects.296 The open source community resists the notion that anyone “owns” the intellectual property of the venture.297 The absence of ownership chafes with some of the strongest arguments in favor of insider trading regulation,298 which rely on the notion of property—that the trader misappropriates the information of another person.299

    Of course, most crypto asset users have not repudiated the state,300 and may welcome enforcement of applicable laws.301 If crypto assets grow to be widely used, their users will increasingly exhibit the same needs and expectations as the population as a whole and thus come to expect similar protections.

    Even users with strong misgivings about the current financial and governmental system may embrace rules that support its alternatives. Gold has long been a mainstay for such investors. Gold has also been the epicenter of widespread market manipulation302 and insider trading.303 Goldbugs do not seem to have abandoned their bullion as prosecutors have sought to reduce market abuse by large banks.304 To the contrary, most gold investors are probably pleased that they can invest without artificial scarcity or volatility. Here, the cops help the skeptics by ensuring the stability of their backup plan. Sometimes the enemy of one’s enemy is one’s friend.

    Recognition of crypto enthusiasts’ distinctive values may even support regulatory intervention, since market regulation and insider trading law turn on what is material305 and crypto enthusiasts simply care about non-financial aspects of the asset.306 For a buyer of crypto assets with one such outlook, there exists an entirely different set of material information bearing on the desirability of a crypto asset.

    Such a buyer may regret their purchase if they discover that the crypto asset’s developers or miners have taken steps to link the asset with familiar commercial institutions,307 or to assist law enforcement officials in tracking illicit purchases using the asset.308 A seller who knows about a pending compromise on one of these axes may sell the asset without disclosing anything to the buyer.309 Can’t we say that the seller sold without disclosing material non-public information?310

    If crypto asset enthusiasts want assets that exhibit certain properties —such as privacy, independence from states and banks—they must be able to trust that the developers and promoters are working to create and maintain such a product. The law can support that trust by recognizing information about the presence or absence of those properties as material. Insider trading law is a natural way to support these efforts. Just as we don’t let insiders sell stock with knowledge that the company really is not profitable, notwithstanding public pronouncements to the contrary, insider trading law can prioritize non-monetary values of crypto enthusiasts in order to support their efforts to seek acceptable assets, and to reduce the chance that fly-by-night promoters sell products that do not live up to past assumptions.

    D. Prioritizing Fraud and Manipulation

    Whatever the merits and fit of applying insider trading law to crypto assets, some commentators have argued that this is hardly the highest priority for enforcement officials. Instead, fraud and market manipulation are far bigger problems for this asset class.

    There is some appeal to this notion because fraud is indeed widespread. Many crypto assets are complete farces.311 Worse yet, crypto assets seem to be vulnerable to a distinctive form of fraud in the form of a 51 percent attack.312 Likewise, market manipulation is rife,313 including familiar techniques such as spoofing314 and pump-and-dump.315 And the existing infrastructure seems inadequate to address market manipulation.316

    For all the problems with fraud and market manipulation, it is not logical to leave insider trading unaddressed while the “larger” issues are dealt with. For one thing, there are costs to having insider trading law that goes unenforced.317 For another, there is literally no tradeoff between one form of enforcement or another to the degree that private litigation is facilitated. Private plaintiffs will bring whatever civil suits are viable. A private right of action exists for plaintiffs who trade in commodity futures based on crypto assets that count as commodities318 or securities.319 They need not show that they directly traded with the insider so long as they traded contemporaneously.320 Already, private plaintiffs have filed civil suits alleging insider trading in crypto assets.321

    More importantly, there is an intimate link between market manipulation, fraud, and insider trading; they are sister sins. To a great degree, they rely on one another to be effective. Reducing insider trading is a powerful way to reduce market manipulation and fraud.

    Insider trading supports market manipulation because market manipulators pose as insider traders.322 Market manipulators seek to influence market prices, and the surest way to do that is to trick other market participants into thinking that some trader out there knows something non-public about the company.323 If a trader wants a stock price to go up, they can buy in a way that implies they know good news about the stock—other traders will take the hint and bid up the stock too, or at least refuse to sell on the same terms as before. This strategy would not be effective if everyone knew the manipulator had only public information.324 Widespread insider trading makes the manipulator’s bluff more credible and effective. Thus, reducing insider trading (or at least enhancing enforcement) helps to reduce the viability of market manipulation.

    Legal insider trading can also provide cover for otherwise illegal market manipulation. Most courts dismiss market manipulation cases if the manipulator had mixed motives.325 That means that a person who intended to manipulate the market, but can cite non-manipulative reasons to trade, will not be liable for market manipulation. The presence of material non-public information can support that defense. For example, in CFTC v. Wilson, a trading firm profited by aggressively bidding for instruments in a manner that pushed up the price in the “right” direction.326 The CFTC’s market manipulation claim failed because the trader had information that suggested the price should eventually go up, and he was just pushing the price toward the “real” price. In that case, there is no argument that the trader’s knowledge of the real price was linked to non-public information acquired in breach of a duty. He was free to trade on it. The more information is a lawful basis for trades, the more market manipulation will find real or pretextual cover.

    The point is not that there is always a positive relationship between one form of market abuse and another.327 The point is that there is always a deep relationship that makes regulation of one a candidate way to reduce regulation of the other—for crypto assets and elsewhere.

    VI. The Limits of Insider Trading Law

    While there is a rich debate about the extent and contours of federal insider trading law, almost all commentators support penalizing trades undertaken with asymmetric information, at least some of the time. Insider trading law is only one form of information regulation. The law of contracts also imposes on parties the duty to correct some errors of an ignorant counter-party.328 The fundamental question of insider trading law is not what precisely to prohibit, but where the background law of contracts is enough and where an additional layer of federal law should be imposed. Does the law of insider trading in securities provide a model for commodities? Crypto assets? Real estate or fine art? Just how wide is the domain of insider trading law?

    One candidate answer links to existing laws: Insider trading law should apply where an asset is already subject to an extensive disclosure regime.329 But disclosure obligation is not a binary classification. Some securities are subject to far less intensive disclosure obligations than others.330 And many non-securities are subject to reporting requirements, even though these assets have struck most commentators as inappropriate for insider trading regulation.331 More foundationally, the choice to extensively regulate is a choice—perhaps fine art should be subject to federal registration, if only to bring it into the domain of insider trading law. Defining the domain of insider trading law by reference to existing laws at best gives us an internally consistent answer, without any assurance that it is otherwise the right answer.

    Several policies used to justify insider trading law are likewise unsatisfying as limiting principles. For example, concern for property rights in information may justify insider trading in securities, but such a theory doesn’t explain why we currently focus insider trading law on securities. There is informational property to steal in other assets, such as art and real estate, but we do not have a dedicated federal agency devoted to addressing misuse of that information. Likewise, if it is unfair for executives to bring home secrets relative to stock prices, it is also unfair for them to bring home secrets relative to real estate prices, but there is no active prosecution of insider traders in real estate. While fairness, property, and other policies are plausible justifications for insider trading law, they are not promising policies for defining its domain.

    To my mind, it is more productive to begin with first principles. To add another layer of law upon the existing contractual information rules is to do two things: First, it is to penalize more conduct.332 The appropriate contours of federal insider trading law are up for debate, but there is no doubt that if there is any federal insider trading regime it will create new impositions not found at common law. Second, it is to establish a new class of professional enforcers: class action plaintiffs, regulators, and prosecutors. With these two core features in mind, we can ask what properties of an asset make it sensible to have greater restrictions on informed trading and to empower a new class of enforcers.

    Each of these principles is capable of careful examination, but the short answer is this: (A) it makes sense to penalize more conduct where traders that lack special information would take costly self-protective measures if exposed to widespread informed traders, and thus harm liquidity; and (B) it makes sense to empower professional enforcers where the “victims” of informed trading would insufficiently litigate their claims on their own.

    Thus, insider trading law is most helpful when it supplements the common law of disclosure with respect to an asset when most traders would excessively protect themselves prior to trading and insufficiently protect themselves after trading.333 The policies explored here are therefore informed by both market microstructure economics and the economics of enforcement.

    A. Restricting Trades to Improve Liquidity

    How much informed trading is it best for any given market to have? Traders who know material non-public information improve the accuracy of asset prices by expressing their informed views through trading,334 and the possibility of trading profits encourages them to acquire information to begin with.335 To the degree prices are accurate, observers of the prices can make better decisions, such as how to invest or redeploy resources in the real economy.336

    But informed trading also has a cost: Informed traders’ profits come at the expense of uninformed traders. Extensive informed trading can demoralize investors from entering a market at all.337 It can also raise the expected cost of trading. In many markets, rising trading costs is reflected in wider “bid-ask spreads” which are the implied commissions charged by market intermediaries called “market makers.”338 Wide spreads make it hard for investors to achieve their trading goals and they blunt the accuracy of market prices.

    Professors Fox, Glosten, and Rauterberg use the tradeoff between price accuracy and liquidity as the main evaluative lens for scrutinizing various types of informed trading:

    How well the market functions can be described largely in terms of its two most important characteristics: price accuracy and liquidity. . . . every type of informed trading has a positive impact on price accuracy and a negative impact on liquidity. But, the ratio of these two impacts and the duration of the price accuracy improvement vary greatly from one type to another.339

    For example, the ratio is “good” for trades by careful researchers of market fundamentals, who bring new information to the market. By contrast, the ratio is “bad” for executives trading just before an earnings report, since the earnings information would be disclosed soon anyway. Fox et al. would allow the former and ban the latter.

    While the price accuracy gains of insider trading differ by type of trade, the liquidity effects do not appreciably do so. If market makers lose money to an informed trader, it does not make any difference how the counter-party acquired their advantage; a bookie loses money against a gambler who knows the outcome of the match, regardless of how the gambler knows. Thus, price accuracy matters in evaluating particular trading practices, but only liquidity matters in evaluating the general domain of insider trading regulation. Assets for which the liquidity harm of informed trading is large should be subject to insider trading regulation in some cases (the details of which must be decided in light of price accuracy effects); assets for which the liquidity harm of informed trading is small should not be subject to insider trading regulations. Therefore, we can decide the domain of insider trading regulation in part based on the liquidity cost of informed trading, prior to any fine tuning in light of price accuracy.

    Consider how this first principle applies to various asset types. Informed trading has a clear cost in terms of liquidity for securities markets. We can be confident that market intermediaries in securities widen their spreads in the presence of informed traders.340 And while it is perhaps hard to prove,341 it is widely believed that investors are demoralized by widespread insider trading and that a relatively level playing field is necessary in order to induce retail investors to pull their money out from under their mattresses—an important social policy goal.

    Similar stories can be told about commodity markets. Trading costs rise in response to expected losses to informed traders.342 It may become more difficult to trade and hedge if sophisticated intermediaries fear that they are in the presence of an insider.343 A similar story can be told about crypto assets, although less definitively.344 The presence of informed traders plausibly affects the reactions of traders. There is no natural user of crypto assets who must trade in them regardless of the market conditions. The goal of many asset promoters of creating a liquid currency is highly dependent upon reducing adverse selection and trading costs.345

    The story is different for other assets, such as real estate and precious art, because the effect of information asymmetries do less to inefficiently alter conduct in the real economy. Buyers of real estate routinely pay six percent commissions to brokers plus myriad other costs.346 Buyers of art routinely pay 50 percent or more in commissions.347 Both transactions are ones where the parties are used to bearing meaningful frictions due to informational imbalances, so the odds are low that a small additional friction will prove debilitating. One reason is that some purchases in this realm are personal rather than financial (the buyer wants to live in the house, or view the artwork), and so may be less elastic as a result. Another reason is that most participants in the real estate and art markets are only occasional participants—they do not buy and sell all day. There are no “dealers” in the real estate world, who stand ready to buy and sell any property at any time.348 Art markets have dealers, but their turnover is also much lower than the rapid-fire transactions of a securities dealer. Taken together, this means that there is no one actor who is constantly penalized by the cost of informed trading and may take socially inefficient self-protected actions as a result.

    Insider trading law prevents informed traders from making gains at the expense of market intermediaries and other traders. Whether this is of great social significance is largely a function of the “victims’” reactions. If they flee the market or charge more for liquidity (as in securities, commodities and crypto assets), then we may be in the domain of insider trading law, and the question then becomes whether there are offsetting price accuracy effects that make one particular type of transaction or another desirable, notwithstanding its cost. If intermediaries and counter-parties don’t greatly adjust their behavior in light of the informed trading (as with art and real estate), then we are likely beyond the domain of insider trading law and existing common law information protections—and freedoms—should prevail.349

    B. Professional Enforcers

    The common law of contract imposes some duties on contractors to disclose information to their counter-party or abstain from trading with them. For example, a seller of real estate may not sell a home with an undisclosed knowledge of a latent defect, just as some sellers of stock may not sell stock with undisclosed knowledge of the equivalent of a latent defect.350 But common law claims are pressed only by the victimized contractor. Insider trading law is enforced by numerous professionals: class action lawyers and government enforcement officials.

    Where some federal theory of liability bars insider trading, securities insider trading law can be enforced criminally by the Department of Justice, or civilly by the Securities and Exchange Commission (“SEC”), a federal agency that oversees the stock markets. Federal laws also allow private civil actions by injured persons arguably harmed by insider traders. For 14e-3 and 10b-5 actions, the trader need not prove they actually bought from or sold to the insider. It is enough that they traded “contemporaneously” with the insider.351 For 16(b), the entity at which the trader is an insider may sue at the demand of any shareholder.

    An additional layer of law is justified to a greater degree if these professional enforcers are justified. Two factors bear on this question: (1) the necessity of expertise, and (2) the value of the asset and asset class.

    Professional enforcers can develop expertise. This expertise makes it easier to enforce the law. But expertise is only sometimes of large value. It is of large value when the subject matter is complex, such that amateur enforcers may bungle things, and when cases are sufficiently alike that there is even some general subject to become an expert in.

    Asset value is the other factor. When an asset class is of great overall value, it is likely worth social resources preventing problems with that asset class. However, single plaintiffs will bring their own case if the value is great to them, too. So, what is required is an asset that makes up a high aggregate value to society, but a relatively low value to the harmed investor. Then it is worth a public enforcer precisely because the victim may just let things go, resulting in underenforcement.352

    These two principles broadly comport with my claims about the domain of insider trading law. To see why, consider the candidate asset classes.

    Securities insider trading is well-situated for professional enforcement.353 The markets are complex. The law, economics, and microstructure of securities and securities trading baffles most law students, to say nothing of investors. It takes professionals to understand finance and to bring cases arising in capital markets. The importance of the subject is also great—the dollar value of securities is staggering, and it represents the industrial policy and retirement prospects of a nation. And it is not just complexity. Insider trading is difficult to detect. It requires costly forensic techniques, such as wiretaps and confidential witnesses, which only the government and sophisticated firms use. Theory and experience teach that professional enforcement is justified if insider trading is to be reduced in securities.354

    A similar story is true of commodities. How many Americans understand the last few minutes of the film Trading Places?355 How many private plaintiffs could spot and prosecute market manipulation in the pork bellies market, or the jargon-laden, international manipulation of Libor?356 Likewise, the stakes are publicly important but perhaps not privately. Although commodities prices sometimes swing wildly, the small changes in price implied by material non-public information is often not sufficient to motivate a victim to sue—in part because her position is hedged and her losses on one instrument are largely offset by another. The long-term harm is one she is in no position to vindicate—widespread breakdown in the value of hedging and speculation instruments.

    Although it is early, it seems plausible that crypto assets fit with the forgoing, such that crypto assets warrant expert enforcers. Crypto assets have a large market capitalization,357 and they may presage a transformative phase in the technological evolution of capitalism.358 But few investors have more than a tiny portion of their savings invested in crypto assets, and those who have lost money in one way or another have little hope of privately pursuing an action. This is compounded by the fact that many promoters and exchanges are located overseas.359 Few users of crypto assets understand finance, economics, and computer science to fully understand the risks they face, the factors which might lead to informed trading, and the path for pressing their rights.

    Real estate and art are different on these two factors. Each property is somewhat unique, making it harder for experts to develop expertise. Traders tend to use the assets they own, giving them a local expertise that may exceed that of professional enforcers.360 And the purchase of a building or precious art is often a serious expenditure for the buyer, who may have incentive enough to litigate unfair dealing.

    VII. Conclusion

    The precise contours of insider trading law are debatable. Should we have an ad hoc rule banning short-swing trades?361 Should the recipient of a gift of information be allowed to trade on it?362 But the conceptually antecedent question involves the domain of insider trading doctrine: For what types of assets should we even be engaged in these questions?

    One common approach to this question is to blithely assume that equity securities stand alone. When some other asset is presented for consideration —corn, bonds, bitcoin—the familiar reply is to deny that insider trading is possible for that asset and deny that the traders in that asset need or want the government’s help. These replies have long dominated discussions of insider trading in commodities363 and traditional currencies,364 but they are now being deployed to exclude cryptocurrencies and other crypto assets from the domain of insider trading law and policy. This Article addressed crypto assets both on their own merits as an important asset class worthy of attention, and also as a token for the broader approach to the paradigm. There is no simple reason to think that crypto assets stand outside of insider trading law and policy. Indeed, many distinctive features of these novel assets make familiar market abuse rationales more applicable than ever.

    But if the line isn’t drawn with common stock, where is it drawn? Should it be a federal crime any time someone buys or sells anything with less than perfect candor? There are costs to insider trading law, whatever its form, and they should only be borne in contexts where they stand some chance of being worth the candle.

    Instead, it is helpful to realize that the core of the matter concerns the marginal contribution of insider trading law to the existing body of contract law: the creation of an extra limit on informed trading and an extra layer of enforcement. These differences are justified where traders react too much in protecting themselves ex ante and too little in protecting themselves ex post. That is to say we should have insider trading law in domains where traders and intermediaries are likely to withdraw from markets due to widespread informed trading—especially in high-volume intermediated markets—and where experts are able to develop expertise in a wide variety of somewhat fungible but complex assets, the value of which to single litigants may be too low relative to the overall social value. Those considerations put crypto assets, securities, and commodities within the domain of insider trading, but leave many other assets beyond.

    Read more: Crypto insider – Krypto-NFTs

    Source: 🔗