Micro to Macro: Causal Micro Evidence for Macro Prevention Models
Project Summary
The aim is to generate robust causal evidence on the economic consequences of poor health. This will be achieved by exploiting rich, linked administrative health and earnings records to obtain quasi-experimental estimates of how medical interventions and major health events affect workers’ employment, earnings trajectories, and firm-switching decisions over time. The focus is on credible identification: natural policy roll-outs, clinical eligibility thresholds, and diagnostic timing shocks will isolate causal effects that standard cross-sectional studies miss. These high-resolution estimates will form the empirical backbone for evidence-based decisions on prevention subsidies, workplace policy, and social insurance design, and will inform the calibration of macro-prevention models.
Why it matters
Poor health is a major contributor to economic inactivity, reduced earnings, and early retirement—yet causal evidence on its economic consequences remains limited. This research helps fill that gap by producing credible, policy-relevant estimates of the value of prevention and the cost of inaction.
- Governments require reliable numbers on the economic cost of chronic disease and the payoff to prevention, yet most existing figures rely on correlations or aggregate averages.
- Accurate micro parameters are essential inputs for agencies such as finance ministries and healthcare technology assessment bodies when assessing the fiscal and welfare implications of health spending.
Why EIT–Oxford is the place
EIT holds secure access to a portfolio of rich data assets that link health and labour‑market histories. Daily interaction with EIT’s prevention, clinical and data‑science teams fosters methodological innovation and rapid policy translation. Oxford’s economics and public‑policy departments add a world‑class setting for labour‑economics training and debate.
Potential Supervisors
- Supervisors are to be confirmed
Skills Recommended
- Graduate econometrics, particularly causal inference methods
- Proficiency in Python, R or Stata for large‑scale data analysis
- Interest in labour‑ and health‑economics questions
Skills to be Developed
- Advanced quasi‑experimental techniques (difference‑in‑differences, event studies, regression discontinuity, instrumental variables) applied to administrative micro‑data
- Translation of empirical findings into policy recommendations and model calibration
- Communication of results to interdisciplinary audiences in health, economics and public finance
University DPhil Courses