The concept originates in contract theory—the recognition that formal contracts cannot specify every contingency, and that parties to ongoing relationships develop shared understandings that govern behavior in the spaces the contract leaves open. Henderson applies the concept to stakeholder relationships, showing that the most valuable firm-stakeholder interactions are governed more by relational contracts than by formal ones. The mentoring relationship between senior and junior employees. The supplier partnership built over years of collaboration. The community trust that permits a firm to operate with social license. None of these can be reduced to contractual terms without destroying the relationship.
AI's challenge to relational contracts is structural. The speed of AI-enabled decision-making exceeds the speed at which relational contracts can adjust. A workforce reduction conceived, modeled, and implemented in weeks does not give the relational contract time to renegotiate. The firm captures the productivity gain. The employees who believed in the implicit agreement—that their development would be supported, that their institutional knowledge was valued, that the firm would not treat them as disposable—discover that the agreement was contingent on conditions that no longer hold. The relational contract is broken not through malice but through the architectural incompatibility between AI-speed optimization and trust-speed relationship maintenance.
Henderson's prescription is that firms deploying AI must make relational contracts explicit—not by converting them into formal contracts, which would destroy their flexibility, but by creating governance structures that give stakeholders voice in decisions affecting them. Stakeholder advisory boards. Expanded disclosure of AI's role in workforce decisions. Compensation structures that reward long-term stakeholder outcomes. These mechanisms do not eliminate the tension between optimization and obligation, but they make the tension visible and governable rather than invisible and destructive.
The term 'relational contract' was introduced by legal scholar Ian Macneil in the 1970s to describe ongoing commercial relationships where formal contract terms are incomplete and where the relationship itself becomes the governance mechanism. Henderson adapted the concept from commercial relationships to stakeholder relationships, showing that the same dynamics—incomplete contracts, trust-based governance, vulnerability to opportunism—operate in the firm's relationships with employees, communities, and other non-commercial stakeholders.
Trust as structural asset. Relational contracts depend on trust, which is not sentiment but an architectural feature enabling coordination under incomplete information and uncertain futures.
Implicit agreements vulnerable to speed. AI-driven decisions execute faster than relational contracts can adjust, making defection structurally easier and trust maintenance structurally harder.
Stakeholder voice as protective mechanism. Embedding stakeholders in decision architectures—not as veto-holders but as voices whose concerns must be addressed—protects relational contracts against AI-speed optimization.
Long-term value from maintained contracts. Firms that preserve relational contracts through technological transitions outperform firms that break them, but only over time horizons that quarterly capitalism cannot perceive.