Trevor I Kiviat. Duke Law Journal. Volume 65, Issue 3, December 2015.
Introduction
On December 26, 2014, three million homes nationwide tuned in to watch the North Carolina State Wolfpack take on the University of Central Florida Knights in the Bitcoin St. Petersburg Bowl—the first of several bitcoin-branded, postseason college bowl games. ESPN’s online presale, held open to sports fans across the nation, involved one catch: prospective attendees could only purchase the tickets with bitcoin. This episode was the first of many that collectively exemplify the mainstreaming of virtual currencies—an atmosphere most recently dominated by the acts of financial players, such as the New York Stock Exchange (NYSE), and state regulators, such as New York’s Department of Financial Services (NYDFS).
Bitcoin discussions largely focus on the technology’s well-publicized growing pains: wild price volatility; fraudulent investment schemes; multimillion dollar hacks; and the infamous Silk Road case—an episode that resulted in a life sentence for Ross Ulbricht, drug kingpin of the deep web, and the indictment of two federal agents. Accordingly, some intelligent and well-respected detractors have called it a “bubble,” and others have gone so far as to call it “evil.” Nevertheless, technologists and business leaders have declared it “better than currency,” citing its promise to lower transaction costs, transform developing economies, and generally “reshape [the financial] system.” Simply put, sensationalism in this area is high. Perhaps this is encouraged by the facts, which read like a science fiction novel, blurring the physical and digital worlds: A pseudonymous inventor releases a cryptographic technology that incentivizes armies of supercomputers to mine digital assets that can be traded for real-world goods and services.
Further, authors almost exclusively focus on bitcoin as a currency system. For example, authors have weighed the costs and benefits of transacting with virtual currencies, considered the sustainability of virtual currencies, and contemplated the application of existing regulatory schemes to virtual currency. Missing from the dialogue is a deeper perspective on the technology.
This Note offers that perspective. Primarily, it expands on contemporary academic literature by highlighting the conceptual distinction between bitcoins (that is, virtual currency) and the “blockchain,” the Bitcoin platform’s key technological innovation. It does this by integrating current research from leading computer scientists and cryptographers. And its ultimate aim is to elevate the legal community’s understanding of blockchain technology and, ultimately, to inform policymakers and practitioners as they consider different regulatory regimes.
In short, the blockchain is a “trustless” technology. “Trustless” means—for the first time in history—exchanges for value over a computer network can be verified, monitored, and enforced without the presence of a trusted third party or central institution. Because the blockchain is an authentication and verification technology, it can enable more efficient title transfers and ownership verification. Because it is programmable, it can enable conditional “smart” contracts. Because it is decentralized, it can perform these functions with minimal trust without using centralized institutions. Because it is borderless and frictionless, it can provide a cheaper, faster infrastructure for exchanging units of value.
Simply, blockchain technology has broad implications for how we transact, and the potential for innovation is hard to overstate. Regardless of one’s opinion on the merits of virtual currencies, financial regulators must develop a better understanding of blockchain technology’s impact potential as they continue to engage in its pragmatic regulation.
This Note proceeds in three Parts. Part I introduces blockchain technology and its most widely understood application: money transfers and payments with bitcoin. First, it explains how blockchain transactions occur and why this technology is highly innovative. Second, it explores bitcoin’s economic properties and situates the currency within the long evolution of monetary technology. Drawing on economic perspectives, it highlights the benefits and drawbacks of a blockchain-based currency like bitcoin. Part I concludes that the technology’s most valuable utility lies beyond bitcoin—in other words, not as a currency but as an exchange medium for digital-asset transfers.
Part II surveys the emerging regulatory landscape, which is heavily premised on the technology’s singular application as a virtual currency. First, it explains the current federal scheme—a patchwork of bitcoin-specific guidance and rulings from the Financial Crimes Enforcement Network (FinCEN), paired with the Commodity Futures Trading Commission’s (CFTC) oversight authority and the Securities and Exchange Commission’s (SEC) enforcement capabilities, which both apply in highly limited circumstances. Next, it explores recent state action—namely, New York’s BitLicense, with special attention to its key provisions and ambiguities. At each layer of regulation, it examines open issues that present uncertainty and opportunity for further clarification.
Part III raises issues presented by blockchain technology beyond virtual currency—beyond bitcoin. It covers applications of special interest to the legal community including more efficient contracts, document and authorship verification, and title transfers. It also explores more advanced aspects of the technology, an understanding of which is essential for sensible policy making in this area. After exploring the vistas beyond bitcoin, this Note concludes by offering thoughts on how caution and restraint might be exercised in the law to facilitate technological and economic growth.
I. The Blockchain, Part 1: Bitcoin, A Blockchain-based Currency
Experiments in currency are as old as commerce and civilization itself. Today, most currencies—the U.S. dollar included—are fiat currencies. Fiat currencies are not backed by physical assets; rather, they are backed by the promise of their issuing government. Commodity monies, by contrast, are backed by a tradable, naturally scarce resource with value beyond its use in trade. Gold or silver, for example, backed the U.S. dollar for much of our nation’s history. This Section explains why bitcoin, the blockchain-based “virtual currency,” does not fit comfortably into either of these traditional categories.
First, this Section answers the fundamental question, “What is bitcoin?” by explaining the lifecycle of a blockchain transaction. Second, it examines the economic properties of an artificial commodity like bitcoin as compared to well-known and widely traded physical commodities and traditional fiat currencies. Finally, it highlights the special properties of this technology—core features that not only enable blockchain-based currencies but also hold vast potential for applications beyond bitcoin.
A. The Blockchain: “Triple-Entry Accounting” on a Transparent, Public Ledger
In the physical world, security requires locks, vaults, and signatures; in the digital world, it requires cryptography, or techniques for securing digital information and transactions. The blockchain is a cryptographic technology. It is the core innovation driving the bitcoin currency system, and it solves an important technological problem. For the first time ever, secure electronic transfers of value can occur without the presence of a trusted third party. By contrast, outside of the blockchain, electronic transfers of value require financial intermediaries—for example, commercial banks, brokerages, or PayPal—to establish trust and security in the transaction. Such institutions establish trust and security by preserving a centralized ledger to track account holders’ balances and, ultimately, vouch for a transaction’s authenticity. Without intermediaries, electronic units of value—dollars, for instance—can be copied and spent twice, just as any digital document can be copied ad infinitum. This “double spending problem” has riddled programmers for decades.
Blockchain technology enables secure electronic transactions without a centralized ledger and without double spending. Instead of a centralized ledger, it makes a collective accounting by distributing a shared (that is, decentralized) public ledger—a complete record of all past transactions on the network. This ledger is the blockchain. When two parties wish to engage in a transaction, they must broadcast it to the entire network, effectively asking network participants to determine its authenticity. The following example illustrates this process.
Party A begins by broadcasting a message to the network signaling the terms of the agreement. For example, “I, Party A, am giving Party B one bitcoin.” Next, Party B accepts the transaction by broadcasting its acceptance to the entire network and asking network participants to determine the authenticity of the transaction. The network automatically validates the transaction—or guards against the threat of double spending—through a “proof-of work” validation system. If the transaction is validated, the ledger is updated and network users’ blockchain records are collectively updated. In other words, once a transaction has been recorded in this transparent public ledger, that transaction cannot be changed after the fact (unless it is matched with a second offsetting transaction).
The proof-of-work validation system is essentially a competition among network participants to validate transactions. Network users participate in this competition by exercising computational power. Under this system, a user’s ability to improperly influence validation—to double spend—is limited by the total proportional computation power he can harness. Users are incentivized to bear the computational costs of validation because successful participants are rewarded with new bitcoin. Accordingly, new bitcoins are said to have been “mined,” with the “[computational] time and electricity that is expended” as “analogous to gold miners expending resources to add gold to circulation.” Eventually there will be nothing left to mine because the total outstanding supply is limited. When that happens, the incentive to validate transactions will likely be transaction fees. Importantly, this is an open-source protocol, meaning open innovation can occur around the technology’s various parameters.
In sum, the blockchain establishes trust between two parties to a transaction through both a decentralized public ledger and a cryptographic mechanism that ensures transactions cannot be changed after the fact. One can easily see why the creator of this technology called it “purely peer-to-peer … electronic cash.” Leaving aside counterfeiting, physical transactions—routine cash transactions, for instance—have never quite suffered from these acute problems of trust and assurance. Yet for the reasons described above, simple two-party exchanges of value over electronic networks could not occur prior to the blockchain innovation.
B. The Economic Properties of a Blockchain-Based Currency
This Section now explores the economic properties of a blockchain-based currency like bitcoin. It examines its basic economic qualities, as compared to commodity money (like gold) and fiat money (like banknotes). It summarizes the key arguments for and against a blockchain-based currency and concludes that, whatever one’s normative views regarding the desirability of such a currency, the technology’s distinctive features indisputably hold potential for the efficient transfer of all sorts of digital value.
Innovation and disruption in the “technology of money” is not new; this competitive landscape has existed for thousands of years. For any technology—be it gold, banknotes, or bitcoin—to be accepted as a monetary standard, it must perform three important functions especially well: it must be (1) a medium of exchange, (2) a store of value, and (3) a unit of account (collectively the functions of money). When a new standard comes along that performs the functions of money better than the incumbents, a platform shift occurs, and the old standard is replaced.
Once upon a time, commodities—shells, grain, and metals—operated as primitive monetary technologies. Among these early prototypes, gold reigned supreme because, of all the naturally occurring elements, its physical properties made it most suitable to perform the functions of money. Despite its first-mover advantage of more than 4000 years, gold was eventually disrupted by the next innovation in monetary technology, government-backed banknotes. Though still a physical technology, banknotes offered streamlined features: portability, divisibility, storability, and fungibility. Soon after, another fundamental shift—this time digital—in monetary technology occurred: electronic deposits and transfers.
1. Bitcoin’s Downside: Blockchain-Based Currencies are a Poor Store of Value. Gold and paper money have worked as monetary platforms because these technologies perform the functions of money especially well. Gold worked as a store of value due to its physical characteristics. The move away from gold was brought on by the realization that commodity money ties a country’s economy to a scarce natural resource, and this can have destabilizing effects. In other words, when Mother Nature controls the supply, shocks can occur that are beyond control. By contrast, fiat currency’s supply—and thus its value—is protected by regulation. It is the only platform recognized as legal tender, the government is obliged to accept it for tax payment, the central bank has monopoly control over supply, and it is often backed by indirect collateral and insurance. These characteristics allow greater price stability. For example, the Federal Reserve can adjust supply to navigate macroeconomic and financial policy issues.
On the issue of value, a blockchain-based currency such as bitcoin is an imperfect substitute for fiat currency in much the same way gold is. The mathematic rules governing the bitcoin mining process are designed to mimic gold. So just as the laws of nature govern the gold supply, the laws of math govern the bitcoin supply. In both cases, supply cannot be adjusted “to deal with recessions or to counteract destabilizing periods of inflation or deflation.” This might explain why the market has experienced wild price volatility.
Over its history, bitcoin’s exchange rate against the U.S. dollar has frequently jumped or crashed over 20 percent (sometimes nearly 50 percent) in the course of a single day. By contrast, over the same period, the U.S. dollar-to-euro exchange rate has never changed more than 2.5 percent in one day. Even a casual observer can recognize that such instability is not a desirable currency trait because its holder’s purchasing power can increase or decrease drastically and suddenly.
2. Bitcoin’s Upside: A More Efficient Medium of Exchange. Though the technology fails as a store of value for reasons described above, the blockchain could play an integral role in the next phase of the financial-technology (fintech) revolution. Given its features, it is a technology uniquely capable of performing several key components of a transaction—recordkeeping, auditing, monitoring, enforcement, or asset custody (that is, escrow)—in addition to facilitating the trade itself. This is important because the global movement of value can be quite cumbersome.
For example, gold and fiat currency have always had high transportation costs, involving security, armored cars, and insurance. In fact, the simple laws of physics limited the Federal Reserve’s original structure; the number and locations of the Reserve Banks are such that “no bank [was] more than an overnight’s train ride from its [Federal Reserve].” These restraints were shattered by the first wave of the digital revolution, in which electronic transfers greatly reduced the cost of moving value.
Yet the movement of value along these electronic systems is still costly. First, moving value—actually clearing and settling a transaction—takes time. For example, on January 26, 2015, the Federal Reserve issued a call to action for all stakeholders in the U.S. payments system to increase end-to-end payment speed, among other things. Currently, the Automated Clearing House (ACH) system supports more than 20 percent of all electronic payments in the United States—these transactions, to a great extent, relate to consumer and small-business transactions. More than $40 trillion moves through the ACH network each year in nearly 23 billion electronic transactions. Nearly all consumer transactions on the ACH network take two to three days. Second, moving money takes money. For example, an estimated $600 billion in principal will be sent in the remittance market in 2015. Companies like Western Union and MoneyGram traditionally provide this service and enjoy an average fee (or “take rate”) of 6 percent, though this rate can run as high as 9 percent. This translates to roughly $36 billion in fees in 2015.
Blockchain technology is uniquely positioned to tackle the problems of both speed and cost. For example, Coinbase, a prominent bitcoin company, provides a service called Instant Exchange. This service facilitates instantaneous cross-border money transfers with bitcoin as the intermediary for a total transaction cost of 2 percent. As applied to the $600 billion principal figure above (today’s remittance market), a potential cost savings of $24 billion might pass through directly to the consumers of such a service.
For these reasons (and many more that are beyond the scope of this Note), the financial-services sector is in the midst of a digital revolution. Of the $23.5 billion invested in fintech ventures between 2013 and 2014, 23 percent ($5.4 billion) was invested in payments technology. As illustrated above, one critical aspect of payments technology is infrastructure. Payments-infrastructure initiatives are emerging in many countries across the world, driven by both public and private actors. Many players—from bootstrapping startups to large, incumbent financial institutions—believe blockchain technology will play an integral role.
In sum, blockchain technology solves an important problem in electronic value transfers. The blockchain does not only move value; it also integrates several components of the trading-clearing-settlement value chain in an elegant, efficient, and mathematical way.
To be sure, these facts neither imply nor foreclose on the desirability of a blockchain-based currency. They simply indicate that blockchain technology should be of interest to any industry engaged in the digital transfer of value. For example, instead of being used as an alternative currency, it might facilitate the transfer of traditional units of value—U.S. dollars or euros for example. In other words, incumbent firms in the payments-and-transfer space can co-opt it to gain efficiencies systems, lower fee structures, and provide more competitive services.
II. The Developing Legal Framework for Blockchain Transactions
Part I explained how money has evolved over time, both as a technology and as a concept. Specifically, it has shifted from a store of value in itself to a medium of exchange. As the role of cash diminishes in favor of electronic deposits and transfers, many wonder about the extent to which blockchain-based currencies will influence the next phase of this global payment revolution. Indeed, entrepreneurial ventures—some backed by considerable human and financial resources—are building a vibrant ecosystem of complementary products and services around this vision. One view, hailing the virtues of a free, open currency market is that transactions in this space should be entirely deregulated. This Part concludes at the outset without further discussion that such a view is neither realistic nor desirable, given a compelling policy interest in preventing abuse and misuse. Examples of such abuses include bitcoin’s potential to facilitate black-market transactions, tax evasion, money laundering, and terrorist financing.
This Part explores the emerging legal framework around virtual currencies and serves as a practical guide for policymakers and innovators trying to both shape and navigate it. Both federal and state regulators have identified some basic risks around blockchain-based currencies and begun staking jurisdictional claims. Policymakers are currently revisiting complex, interwoven regulatory frameworks-primarily banking laws, commodities laws, and securities laws—to shoehorn the technology into existing frameworks and consider where new ones might be appropriate. This Part presents a patchwork that is continuing to emerge, with special attention on areas posing uncertainty for innovators.
A. Federal Regulation of Blockchain-Based Currencies
No comprehensive federal regulation exists for virtual currencies. Many government bodies—specifically, FinCEN, the Internal Revenue Service (IRS), SEC, CFTC, and Consumer Financial Protection Bureau (CFPB)—have offered guidance and taken limited action. This Section summarizes the most significant federal developments to date—FinCEN’s guidance, administrative rulings, and enforcement against Ripple Labs, Inc. (Ripple)—and explains the likely implications for innovators. Finally, it notes the limited scenarios in which the other agencies have jurisdiction over blockchain activities.
1. FinCEN Guidance, Rulings, and Enforcement. Under the Bank Secrecy Act (BSA), banks and other financial institutions are subject to various registration and recordkeeping requirements. All “money service businesses” are required to register with the Department of the Treasury and develop anti-money-laundering and customer identification programs. In March 2013, FinCEN extended these rules to cover certain participants who transact in “convertible virtual currencies.” It defined this term to include any medium of exchange that “operates like a currency in some environments,” and “has an equivalent value in [or acts as a substitute for] real currency,” but does not have “legal tender status in any jurisdiction.”
Under FinCEN’s guidance, “exchangers” and “administrators” are possibly subject to regulation. Exchangers are persons or businesses that exchange virtual currency for real currency, funds, or other virtual currency. Administrators are persons or businesses engaged in the business of “issuing (putting into circulation) a virtual currency” who also have “the authority to redeem (to withdraw from circulation) such virtual currency.” An exchanger or administrator becomes a “money transmitter” subject to these registration and recordkeeping requirements when they either “accept and transmit” convertible virtual currency or “buy or sell” convertible virtual currency. “Users” are explicitly carved out.
In 2014, FinCEN issued four rulings under this guidance that, together with existing BSA laws, provide some key insights. First, any blockchain transaction is likely a virtual-currency transaction, because even nonfinancial uses require a de minimis amount of currency (that is, a fraction of a penny of bitcoin). However, such activity must also be performed by an “exchanger” or “administrator” to trigger BSA requirements. End users, such as merchants or consumers, are likely to be exempted.
Second, a user who mines virtual currency (miner-user) is not a money transmitter, even if he uses the bitcoin to purchase goods and services. Further, miner-users converting virtual currencies to real or other virtual currencies are not subject to BSA requirements, so long as their conversion is for personal use. Therefore, miner-users should not be seen as money transmitters subject to the BSA’s registration and recordkeeping requirements unless they are selling bitcoin as a business.
Third, a company that mines virtual currency (miner-company) is not a money transmitter in certain instances. Specifically, miner-companies are not money transmitters when convertible virtual currency is used (1) to pay for goods or services, (2) to pay debts previously incurred, (3) to make distributions to owners, (4) to purchase real or other virtual currency specifically for any of the previous three purposes, or (5) for the company’s own investment account.
Fourth, a company is an “exchanger” regardless of whether it acts as a broker (by matching two simultaneous, offsetting transactions) or as a dealer (by transacting on its own account). At least three U.S.-based exchanges have shut down in the wake of this guidance.
Finally, two important exemptions (that predate both the guidance and the rulings) carve out certain activities from the definition of money transmitter: the “integral” exemption and the “payment processor” exemption. First, BSA legislation provides an exemption for entities that accept and transmit funds “only integral to the [entity’s] sale of goods or the provision of [other, nonmoney transmission] services.” In other words, ordinary merchants and service providers who merely accept bitcoin as a convenience to customers are not money transmitters. Second, BSA legislation provides an exemption for any entity acting as a “payment processor to facilitate the purchase of … a good or service through a clearance and settlement system by agreement with the creditor or seller.” One condition necessary for this exemption is that the entities operate only through clearance and settlement systems that admit BSA-regulated financial institutions. Accordingly, bitcoin-based payment processors will have a difficult time availing themselves of this exception because the virtual-currency leg of the transaction will always settle on the blockchain—a system that inherently allows participation by non-BSA-regulated members.
On May 5, 2015, in its first civil enforcement action against a virtual-currency business, FinCEN announced a $700,000 fine against Ripple and a simultaneous settlement agreement. Ripple was selling XRP, a virtual currency similar to bitcoin, that it designed for the purpose of creating a real-time settlement infrastructure. In its negotiated settlement with the U.S. Attorney’s Office in the Northern District of California, Ripple admitted to violating several BSA requirements in its “exchange” and “transmission” of XRP for fiat currency. Though Ripple had registered its subsidiary as a money-services business in accordance with FinCEN’s guidance, it sold XRP for several months without a proper anti-money-laundering (AML) program in place, failed to designate a compliance officer, and did not solicit an independent review of its practices and procedures.
Two lessons can be learned from FinCEN’s enforcement against Ripple. First, FinCEN is clearly taking a hard stance, per its 2013 guidance, that AML programs are a necessity from the very moment a business begins “exchang[ing]” or “transmi[tting]” customer funds. Second, distributed-ledger businesses that operate outside of the traditional Bitcoin blockchain will not escape FinCEN’s scrutiny.
2. CFTC Jurisdiction over Bitcoin Derivatives and Market Manipulation Oversight. As noted above, blockchain-based currencies share some economic properties with commodity money, and legal definitions support their characterization as a commodity in some instances. The Commodity Exchange Act (CEA) broadly defines a “commodity” to include “all services, rights and interests … in which contracts for future delivery are presently or in the future dealt in.” Accordingly, the CFTC has jurisdiction over derivatives contracts related to interests not traditionally thought of as commodities—Treasury securities, stock-market indices, and currencies, for example. Under this analysis, the CFTC concluded that bitcoin and other virtual currencies are “properly defined as commodities.” And in September 2014, the agency oversaw the launch of the first bitcoin swap execution facility (SEF).
Bitcoin derivatives—for example, a swap contract pegged to the U.S.-dollar-bitcoin exchange rate—are exotic instruments at this stage. The more pressing question, then, is the extent to which the CFTC can exercise jurisdiction over spot-market transactions under its anti-manipulation authority. In other words, the CFTC has enforcement authority over spot transactions in certain instances because spot-market manipulation can affect derivatives market prices. Thus, in certain cases the CFTC may regulate bitcoin pursuant to its anti-manipulation rules. While manipulation oversight would bring some regulation to the spot market, one issue is whether manipulation oversight alone is sufficient, even under the broad anti-manipulation rules of the Dodd-Frank Wall Street Reform and Consumer Protection Act (Dodd-Frank).
Dodd-Frank extended the CEA’s anti-manipulation rules to cover swaps and clarified that “manipulation” under the CEA includes not only “actual manipulation” but also an intent-based “attempted manipulation.” This new authority was first exercised in CFTC v. Atlantic Bullion & Coin, Inc. In Atlantic Bullion, the CFTC brought a civil action against the coordinators of a Ponzi scheme involving spot-market silver contracts. Over an eleven-year period, the defendants fraudulently sold silver contracts in a nationwide scheme. The defendants never supplied any silver; instead, they misappropriated all the funds and issued false account statements. Under a similar analysis, the CFTC could bring investor-protection measures to the spot market for blockchain-based currencies and derivative products.
B. State Regulation of Blockchain-based Currencies
On June 3, 2015, New York’s Department of Financial Services issued its final “BitLicense” framework for regulating “virtual currency businesses.” Over a period of almost one year, BitLicense went from its initial proposal to reproposal to final rule. The process gave rise to two comment periods that elicited thousands of letters expressing a wide range of opinions. And although it is too early to draw any empirical conclusions on BitLicense’s long-term market impact, it has certainly raised the cost of entry for certain participants and will likely pave a smoother path to integration with the established banking system.
New York’s regime covers most business activities involving (1) “virtual currencies,” defined to include decentralized blockchain-based currencies, and (2) New York or New York customers. Much of the uncertainty around BitLicense lurks in its protracted definition of “virtual currency business activities,” which breaks down into five major prongs: (1) transmitting virtual currency; (2) holding virtual currency on behalf of others; (3) buying and selling virtual currency as a customer business; (4) providing exchange services as a customer business; and (5) controlling, administering, or issuing virtual currency.
First, the “transmission” prong presents some uncertainty in the statutory language itself. For example, the definition includes “the transfer, by or through a third party, of Virtual Currency from a Person to a Person.” Imagine a business that simply transfers virtual currency internally between proprietary accounts. In such a case, the transmission prong rightly is not triggered because the business does not interact with any third parties. However, what if that same business also transfers virtual currency to third parties, but not for goods or services—for example, to pay dividend distributions or salaries? The statutory language does not resolve whether the business, by virtue of that fact alone, must carry a BitLicense.
Another wrinkle in the “transmission” prong is the explicit exception for transactions “undertaken for non-financial purposes” that do not involve “more than a nominal amount” of virtual currency. “Non-financial” is not a statutorily defined term. Blockchain technology can be used in a number of ways that are clearly “non-financial”—for example, to facilitate identity verification, digital-document verification, or peer-to-peer transfers of digital assets. In other cases, however, it is less clear whether this exception applies. For example, how would a smart contract transferring a right to payment from financial assets using a nominal amount of virtual currency be treated?
Second, the “holding” prong presents uncertainty with respect to its scope. Though the draft language includes the word “securing,” that word is absent from the final rule. “Securing” virtual currency likely refers to multi-signature (“multi-sig”) transactions. Multi-sig transactions involve more than two parties. For example, a two-of-three multi-sig transaction is a transaction between three parties that requires the approval of two parties prior to settlement. One implication of this feature is cryptographic escrow. For example, Party A and Party B enter a contract with payment provisions contingent on an objectively verifiable event. They enlist Party M as a mediator who will sign the transaction in favor of the appropriate party upon the occurrence or nonoccurrence of such event. Removing Party M from the scope of this prong is probably appropriate because Party M never actually takes custody of the assets.
Third, the “buying and selling” prong presents uncertainty in the statutory language. Specifically, this prong is triggered by the buying and selling of virtual currency “as a customer business”—a phrase that, read broadly, could likely encompass a wide range of activity. The best way to view this prong seems to be that it refers to buying virtual currency from customers and selling virtual currency to customers on a principal or agency basis. Under this interpretation, sales of virtual currency to third parties that are not part of the customer-facing business should fall beyond the provision’s scope.
Both the fourth and fifth prongs (that is, the “exchange services” and “controlling or administering” prongs) overlap with FinCEN’s definitions of “exchangers” and “administrators” under FinCEN’s 2013 guidance. Likewise, the same analysis that applies under FinCEN’s 2013 guidance would apply to covered activities under both prongs. Miners and creators of decentralized virtual currencies likely would be excluded under the same reasoning, assuming their activities extend no further.
Lastly, two exemptions are worth noting. First, the “merchant/consumer” exemption is fairly straightforward. Like FinCEN’s 2013 guidance, it carves out merchants or consumers who use virtual currency solely for purchasing or selling goods or services, or solely for investment purposes. Second, a more ambiguous “software developer” exemption applies to individuals and businesses that engage solely in the development and dissemination of software. NYDFS has consistently asserted that it is regulating financial intermediaries, not software developers. However, the line between the two may not always be clear.
Consider a business that develops wallet software—mobile applications that allow users to view and manage their virtual-currency balance. On one hand, the developer does not take custody of the user’s virtual currency at any point, and it does not transmit or exchange virtual currency. Instead, it simply provides the user with a blockchain access point. On the other hand, the software stores the user’s private key—the secret mathematical code necessary for the user to access his holdings on the blockchain. This weighs against the exemption’s application, because access to a user’s private key is the functional equivalent of access to the user’s holdings tied to that key. Accordingly, a security compromise in the wallet software could cause users to lose all or part of their virtual-currency holdings.
In light of the prior analysis, it seems fair to say that the law will have at least two short-term consequences. First, it will raise the cost of entry for market participants by mandating various programs—cybersecurity, consumer protection, financial reporting, and AML. Indeed, many businesses have already chosen to exit New York, citing total compliance implementation costs between $50,000 and $100,000. Second, the certainty of licensure decreases legal risk of companies operating in this space, so a smoother path will likely emerge for blockchain businesses to integrate with the established banking system.
III. The Blockchain Revisited: The Shape of Transactions to Come
This Part builds on the explanation of blockchain technology set forth in Part I and illustrates why regulations designed to “broadly construe” the definition of “virtual currency” may unintentionally engulf an entire realm of activities. First, it explains the concepts of “scripting” and “sidechains”—innovations that could spawn additional applications for blockchain technology. Second, it surveys current research and experimentation at the cutting edge of cryptography and computer science that could impact commerce and on a similar order of magnitude as the Internet did. It closes by circling back to themes raised in Part II, exploring the challenge that regulators face as they seek to understand this technology.
A. The Blockchain Revisited: Scripting and Sidechains
Potential applications of blockchain technology are not limited to money transfers and payments. At its core, this protocol facilitates more than the exchange of “bitcoins”; it facilitates the exchange of value. Part I established a series of important mathematical rules that govern the network. Fundamentally, transactions have a three-part structure: (1) Party A sends a message to the network declaring the transaction; (2) Party B accepts the transaction by broadcasting its acceptance; and (3) the network participants verify the transaction’s authenticity. To be sure, this basic structure was designed for transferring ownership of bitcoins. But when people send and receive bitcoins, those bitcoins are best thought of as containers for value. Like a digital envelope, these containers can carry “coins” across the network; but they can also transmit richer forms of information, holding promise for many compelling applications beyond bitcoin.
A typical transaction follows a simple script—a set of instructions—that adheres to the three-part structure described above. If the script were amended to contain additional conditions, users could engage in more sophisticated transactions. For instance, consider that Party A and Party B may want to add a fourth condition to that script structure: they only want the transaction to occur at a certain time, or upon the occurrence or nonoccurrence of a conditional event. Many possibilities branch out from this basic idea, and it has sparked much discussion around “smart” contracts.
As a practical matter, developers cannot currently implement scripts like this in bitcoin transactions because protocol amendments require a majority consensus. Similar to a corporate charter, default rules are easy to establish at the outset and much harder to change later on. This fact, paired with the open-source nature of the Bitcoin platform, has inspired dozens of “altcoins,” or alternative-utility iterations on blockchain technology. In other words, developers with a novel vision for the ideal blockchain parameters set their own rules at the outset, according to a desired set of economic properties. Some examples are Litecoin, a platform similar to Bitcoin but with faster transaction confirmations, an ideal feature for high-volume merchants; Viacoin, a “notary” platform that timestamps, transfers, and verifies ownership of documents; and Storjcoin, a platform much different from Bitcoin that allows for a decentralized cloud storage system.
Despite the excitement of this unbounded innovation, a system of parallel blockchains is inefficient and undesirable. They also pose significant risks to the sustainability and goodwill of the blockchain experiment. Although a full discussion of these risks exceeds the scope of this Note, they generally fall into one or more of the following categories: problems of initial distribution and valuation, liquidity shortages, adverse network effects, market fluctuations, fragmentation, security breaches, pump-and-dump market games, and plain fraud. The good news, however, is that a recent development has shown these “worlds” of alternative-utility blockchains can coexist without the exchange-rate risk and other factors that make the current altcoin system unworkable.
In October 2014, a group of leading developers introduced the concept of “sidechains.” Unlike altcoins, which require users to leave the Bitcoin platform, exposing them to significant risks, sidechains are blockchains that are interoperable with one another and, most importantly, interoperable with the Bitcoin blockchain. By integrating with Bitcoin’s blockchain, sidechains provide the benefits of altcoins without the accompanying risks. Such purpose-specific scripting will encourage further innovation by allowing for a network of “distributed trust systems.”
B. Decentralized Smart Contracts and the Shape of Transactions to Come
Sidechains and scripting are changing how people think about blockchain technology. One broad area of innovation around these features is decentralized smart contracts. Smart contracts are “computer protocols that facilitate, verify, execute and enforce the terms of a commercial agreement.” This concept is not new and is not unique to the blockchain. One primitive example is digital rights management (DRM), a technology developed to fight copyright infringement. DRM technology essentially embedded U.S. copyright law into digital files by limiting the user’s ability to view, copy, play, print, or otherwise alter the works. In other words, digital audio files encrypted with DRM technology were not subject to the double-spending problem because they contained a basic smart contract, one that referenced a centralized network, (that is, Apple’s server programmed to enforce the iTunes Store Terms and Conditions).
The blockchain enables decentralized smart contracts—in other words, smart contracts that leverage a secure public ledger as an enforcement mechanism. In contrast to the iTunes example, these contracts do not rely on a third-party institution or server for centralized recordkeeping and enforcement. Because blockchain transactions are programmable and self-enforcing, parties might use smart contracts to design contractual relationships that are automatically executed without the additional costs of monitoring or enforcement.
This fact is significant. Intermediaries typically establish trust and reduce risk between counterparties to a transaction. But with decentralized smart contracts, parties may transact at arms length, with total strangers, without the worry of fraud, and without the cost of third-party enforcement (that is, recordkeeping costs, mediation costs, and other administrative and operational costs). In other words, decentralized smart contracts allow for new markets to develop: disintermediated contract markets in which parties do not have concern for counterparty risk.
Consider a smart-contracts market for futures trading. Smart contracts in this market would be simple for two reasons. First, futures agreements involve objectively verifiable conditions about the state of the world—for example, the price of crude oil at a given time on the New York Mercantile Exchange. And second, futures agreements are highly standardized to ensure that contracts can be easily traded and priced. Such an agreement would be self-monitoring and self-enforcing through a combination of scripting, multi-sig, and oracles, systems set up to monitor off-blockchain information and data that is essential to the effective execution of the smart contract’s terms.
In sum, the technology’s potential to lower transaction costs with respect to contracting and transferring title to physical and personal property should generate special interest in the legal community. To be sure, there are challenges. First, the task of encoding the legal subtleties and nuances that underlie even the most basic contract poses significant programming challenges. And second, it is not clear whether and how smart contracts fit within the legal frameworks of the Uniform Commercial Code and general common law. Although an extended discussion of these two issues is beyond the scope of this Note, their serious analysis would add much to this nascent field.
Conclusion
Blockchain technology is adaptable and policymakers must view it as such. Regulation designed to mitigate the risks of such a powerful technology should be encouraged. However, policymakers should exercise caution and precision in tailoring the scope of regulation. As illustrated above, blockchain technology has utility beyond transmitting value in the traditional money-transmitter sense. Regulation aimed at the blockchain’s money-transfer and payment functionalities must not create an unintentional chilling effect on this second category of functionalities.
States should monitor New York’s BitLicense experiment and consider the issues raised in this Note as they consider their own models. For example, the NYDFS has recognized that BitLicense is intended only to apply to financial intermediaries. This Note highlighted some ambiguity around “nonfinancial” use language. Further, depending on particular alternative applications of blockchain technology, some additional guidance and regulation may need to occur outside of the BSA and state banking frameworks. For example, smart contracts that enable equity crowdfunding should fit squarely in the domain of federal securities law, triggering registration and disclosure requirements and subjecting participants to SEC enforcement rules. In other words, policymakers must carefully define the specific activities that they seek to regulate. A basic understanding of the concepts set forth in this Note would be a strong starting point. To borrow from technologist Mark Stefik’s words on the Internet, blockchain technology can support different kinds of dreams: “We choose, wisely or not.”