Discussion: (0 comments)
There are no comments available.
A public policy blog from AEI
The latest on technology policy from AEI, published daily.
More options: Share,
All contracts are incomplete. That is, humans can’t foresee all possible future outcomes, so it’s impossible to identify and specify terms addressing every possible contingency. If there is too much uncertainty, otherwise-beneficial trades might not occur. The welfare-enhancing solution to incompleteness is to accept that a contract is imperfect but agree to terms (e.g., arbitration) in the event that unanticipated outcomes arise. The harder it is to specify what the terms are, the more discretion must be allowed to the arbitrator. For some extremely hard-to-specify contracts, such as employment agreements, there may be no terms specified. Rather, a set of principles guides the parties’ behaviors, and significant discretion (albeit guided by precedents in the case of courts) is granted to the arbitrator. Such flexibility is widely agreed to enhance efficiency.
The latest innovation — or challenge — to contracting sits in blockchain distributed ledger technology. While cryptocurrencies have been the first high-profile applications to make use of blockchain, an even greater potential for revolutionizing commercial interactions comes from the technology’s ability to support “smart contracts” — computerized transaction protocols that automatically execute the terms of an agreement between two participants. The Ethereum blockchain is an example. Drawing on the Turing complete features of its proprietary programming language, smart contract code — instructions that will self-execute at a future date when specified preconditions are met — can be written to operate on the subsidiary Ethereum Virtual Machine runtime environment, which also hosts the platform’s fundamental consensus mechanism.
Disintermediating the courts?
Many different uses for smart contracts have been suggested — for example, in international trade, where payment from a buyer to a seller can be placed into escrow and released when proof of dispatch has been provided (e.g., by a “smart sensor” in the courier’s system recording the package leaving the seller). However, a particular problem arises, as a blockchain transaction is irreversible once enacted. What will happen if the goods aren’t those expected by the buyer or they are no longer needed?
In the real world, such events can be attended to using the courts or other agreed-on arbitration processes. It has been suggested that the power of smart contracts is such that code can also be created to automate dispute resolution. For example, Kleros, operating on the Ethereum blockchain, uses “blockchain technology and crowdsourced jurors to adjudicate disputes in a fast, secure, and affordable manner.” The buyer and seller, when agreeing to terms of the original transaction, could opt to invoke a Kleros decision in the event of a dispute. The Kleros decision cannot reverse the original transaction, but it can generate another preprogrammed transaction to compensate the adjudged loser at the expense of the counter-party.
While the Kleros solution to this problem appears intuitively appealing, it raises some important questions about the realities of real-world contracting, the extent to which intentions can realistically be captured in code on a blockchain system, and whether moving toward such arrangements may have implications for the nature of contract law.
The only certainty is uncertainty
The difficulty with smart contracts is that all the terms and contingencies must be specified in advance and coded into the blockchain. While it may be possible to specify contingencies for expected events (e.g., the moral hazard actions that occupy game theorists developing the smart contract algorithms and code), many outcomes can’t be predicted and are beyond human control. Code cannot be developed to address events that will happen after participants agree to a contract. As the smart contract cannot be stopped, the best participants can hope for is a process that minimizes losses and distributes them “fairly” — given that events may have changed what is deemed fair.
Real-world courts, run by humans guided by principles rather than bright-line rules, have for centuries been trusted with applying their wisdom to these types of matters. Their decisions are among the hardest things to render into simple rules and procedures, because of the huge variety of factors that must be taken into account. Systems such as Kleros propose crowdsourcing arbitrators to rule on straightforward computer-generated scenarios and evidence, with a majority response to a range of likewise computer-generated questions and remedies prevailing. But how can they do so in respect of contingencies that cannot be anticipated or when events make previously unimaginable options available? This appears to be one of the biggest weaknesses of smart contracts and likely will discourage their use for all but the most certain of outcomes.
Others, however, have argued that smart contracting and other algorithm-based applications would be made more desirable if the law itself was structured less around principles and used more binary bright-line “yes/no,” “in/out,” and “acceptable/unacceptable” rules. Primavera De Filippi and Samer Hassan describe this as a move from “code is law” to “law is code” and argue that there is already evidence of laws that are more “regulatory” (prescriptive) than principles-based in nature. That is, the only acceptable terms are those specified by the regulator ex ante (and hence able to be captured in code), rather than more “competition law-based,” where freedom to engage in a wide range of different terms is possible, with violations identified and acted on ex post and ruled on by principles-guided courts. If this eventuates, then it appears likely that incentives for contractual innovation will reduce. Outcomes will be more predictable, but not necessarily more welfare-enhancing.
Back to the center
Ironically, greater use of (innovative?) decentralized smart contracts appears contingent on a degree of standardization more reminiscent of preordained centralized control than a flexible system (e.g., competition law) based on ex post resolution by human-led courts.
Note: Bronwyn Howell and Petrus Potgieter are convening panels discussing issues of blockchain technology and governance at the forthcoming International Telecommunications Society’s 30th European Regional Conference in Espoo, Helsinki (Finland), on June 17 and the Workshop on the Ostrom Workshop at Indiana University, Bloomington (USA) on June 19.
There are no comments available.
1789 Massachusetts Avenue, NW, Washington, DC 20036
© 2019 American Enterprise Institute