Disentangle price effects from congestion and queueing in observed participation responses

Develop empirical strategies and data that incorporate direct measures of waiting time, queue length, and real‑time availability to determine the extent to which observed participation responses attributed to posted time‑of‑use prices are instead driven by congestion avoidance and queueing behavior.

Background

The paper notes that station‑level participation can be influenced by real‑time congestion and queueing, which are not modeled explicitly. Such dynamics can mimic price sensitivity on the extensive margin, confounding interpretation of reduced‑form estimates based solely on posted prices.

The authors emphasize that without direct operational measures of congestion and waiting, it is impossible to separate price‑driven behavior from congestion‑avoidance behavior, leaving this mechanism untested in the present data and motivating targeted data collection and modeling.

References

Queueing behavior can generate patterns that look like price sensitivity in participation even when the main driver is congestion avoidance. Without direct measures of waiting time, queue length, or app-based availability information, these channels remain untested.

— The Effectiveness and Limits of Time-of-Use Pricing in Public EV Charging Networks  (2603.29223 - Xiao et al., 31 Mar 2026) in Section 7 (Limitations and Outlook)