Perfectly Still
Stability and Policy Response Capacity
Research question
How does policy history change the response capacity available to future interventions?
Abstract
Earlier interventions can change the response to a later policy. This paper studies that response as a state-dependent map from current instruments to future targets. Its evolution reflects entry, exit, and changes among units already reached. Adoption and effective-cap models provide mechanisms through which policy history changes the map. Sequential randomization and factorial assignment identify distinct history and configuration contrasts under their stated designs. In the myBPmyLife trial, the fitted current diet-notification response is 1.75 percentage points lower at the 90th than at the 10th percentile of fourteen-day assigned history, conditional on the recorded pre-state and time controls. A separate factorial experiment shows how an app-limit assignment changes the response to a temporary use-reduction bonus. Rank and finite-action calculations describe the directions available for the targets and horizons represented in each experiment.
In one sentence
A system can look stable today while its ability to respond to tomorrow’s intervention still depends on the policies it experienced before.