From its very early days, one of the main areas where blockchain enthusiasts argued there was potential for radical innovation was in the ability of the technology to house smart contracts. The Ethereum network was designed with this in mind: to be a distributed ledger with a ‘Turing complete’ virtual machine on top of it. With this you could great a program that was effectively a contract performance system — triggering payments when information was received regarding the state of the world or the performance of other contractual obligations.
To date, the use of smart contracts has been few and far between. There are some applications which are substitutes for escrow services (that is, where payments — in tokens — are made contingent upon certain events). There are others were the blockchain is used to generate crowd controlled lists (e.g., token curated registries). Suffice it to say, this use case has yet to be borne out.
Given the potential, I have taken a close look at this area and have just completed a paper — “The Fine Print in Smart Contracts” — that takes a rigorous economic approach to what smart contracts are and where blockchain technologies may allow them to do more than traditional arrangements. The primary issue that needs to be tackled is the “Fundamental Smart Contract Challenge.”
The challenge for smart contracts comes with regard to performance obligations that need to be verified — that is, to provide hard evidence that an outside party would be able to use to confirm contract performance. In some situations, hard evidence — or its equivalent — can come when the party receiving the obligation confirms that the obligation has been performed. But for the most part the challenge arises because a party performing an obligation has an incentive to claim an obligation has been performed while those receiving it may have an incentive to claim it has not been performed; this is especially the case if their own obligations (e.g., making a payment) is contingent upon the obligation being reported as performed.
In other words, smart contracts necessarily have to deal with human decision-makers in the real world and so have to respect their incentives. If a smart contract is executed based on a human action then it had better be the case that the human has the incentives to perform that action rather than pretend to do so.
So where might there be hope for smart contracts on the blockchain? In the following video I explain that starting from first principles, the blockchain does offer the potential to create good incentives through the use of tokens but that the mechanisms that will allow this to occur are still governed by the laws of economics.
That video leaves open what mechanisms might work. There is certainly potential for real innovation in this space. But, for the most part, people have not been doing in this in a rigorous way that respects the knowledge we already have on the subject in economics.
In the paper, I suggest a way of using ‘subgame perfect implementation’ (a decades old idea in economics) on top of the blockchain to solve the international trade problems outlined in the video. That situation is where a buyer and seller are separated by geography and countries and so cannot efficiently rely on legal institutions to resolve contractual disputes. The blockchain allows a smart contract to be written so the buyer pays the seller if a product is delivered. However, without an additional mechanism, it will not get the seller to ship the product expected rather than some inferior substitute.
The way the mechanism works is that if there is a dispute over product quality, it automatically generates costs for both the buyer and seller. In effect, at a cost, the buyer can ask for a refund but the seller has the opportunity to counter with a discounted price instead. Whoever ends up being in the wrong (the seller if the return is made or the buyer if a discount is accepted) ends up paying the most of the dispute costs. The fear of avoiding those costs causes both parties to perform according to their expected obligations. This is done with no external legal mechanism using only the token/escrow type functions that the blockchain is good for. It is a neat mechanism and the reason why it has not been widely used to date is that it was waiting for something like the blockchain to come along.
In summary, the missing element in smart contracting is likely economic innovation. It is unclear, at this stage, whether that can work out but it does present some interesting new opportunities.