---

Many of our seminars and public events this year will continue as in person or as hybrid (online and in person) events.

Please check our website listings and Twitter feed @STICERD_LSE for updates.

Unless otherwise specified, in-person seminars are open to the public. Please ensure you have informed the event contact as early as possible.

Those unable to join the seminars in-person are welcome to participate via zoom if the event is hybrid.

---

About this event

A large literature in health economics and other fields (e.g., education, urban economics) documents aggregate differences in the impact of physicians (teachers, neighborhoods) on randomly assigned patients (students, residents). Whether these aggregates inform an individual?s choice and welfare, such as a physician?s treatment of a specific patient, is not well established. We show theoretically that Counterfactual Homogeneity (CH), a condition we formulate, is necessary and sufficient for aggregate estimates to apply to specific patients. CH requires that counterfactuals can be mapped by a function that is homogeneous across all patients. Testing CH requires estimating the causal effect of different physicians on specific patients, which is difficult with standard datasets and across-subject designs. Using a novel dataset on standardized patients in India, where we observe such patient-level differences in practice, we test CH directly. CH is frequently violated across cases that differ in diagnostic uncertainty. For example, it is violated across patient gender when diagnostic uncertainty is low, but not when it is high. Welfare estimates of policies that reallocate physicians to patients based on aggregate estimates are severely biased when CH is violated. Finally, we provide a statistical test based on over-identification restrictions that lets researchers test CH without standardized patients.