Provable transactions: Exploring the boundaries of trust in smart contracts on blockchains
Trust is a precondition for successful transactions. Yet, it becomes dispensable if a party can ex ante comprehend and ascertain the mechanism by which its counterpart will fulfill its contractual obligations; in that case, the focal party can obtain proof and trust is replaced by certainty. We refer to such transactions as provable, and extend the theory of trust by developing a three-layer model of initial trust formation for settings involving provable transactions. We test our model using a sample of 228 smart contracts deployed on the Ethereum blockchain. Smart contracts are immutable, publicly visible, and automatically executed computer protocols which allow to ex ante verify the outcome of a transaction and thus hold the promise of facilitating provable transactions. With a moderated regression model, we test which trust layer is responsible for trust formation and on which factors this formation is contingent.
Our findings suggest that blockchain and smart contracts do not remove the necessity of trust from transactions but may induce a shift in trust assessments from characteristics of the offering company towards integrity of the smart contract itself. This effect is especially prevalent if transaction risks are high.