S-Frame Vs. I-Frame Interventions: The Transformation-Robustness Trade-off in Behavioural Policy

This paper evaluates the risk–benefit profiles of individual-level (i-frame) and system-level (s-frame) behavioural interventions using an agent-based model (ABM). The i-frame/s-frame distinction refers to whether policy targets individual behaviour change (e.g., nudges, information campaigns) or systemic structures (e.g., regulations, network-based seeding). We formalise and test three policy-relevant risks: (A) structural misspecification risk (errors in the network used for targeting), (B) heterogeneity risk (adverse or weakly responsive subpopulations), and (C) shock risk (temporary dampening vs. backsliding of behaviour). For each dimension, we track four outcomes: adoption levels achieved, diffusion speed, cost-efficiency, and downside risk (measured by the share of runs ending below 10% adoption). Two findings stand out. First, when structural assumptions are approximately correct, targeted s-frame seeding ignites cascades and is markedly more transformative and cost-efficient than i-frame benchmarks. Second, the same reliance on social spillovers creates fragility:mis-targeting and backsliding shocks substantially reduce performance, whereas i-frame designs deliver smaller but steadier gains and degrade more gently. The results imply a portfolio approach: measurement-first targeting and staged rollout to earn the right to scale s-frame power, coupled with i-frame backstops and simple shock absorbers to keep systems in a safe-to-fail corridor. The framework provides actionable guidance for choosing and sequencing policy levers under uncertainty while making risk explicit.