ADS & AMDS Webinar | Modelling Economic & Health Effects of COVID-19 Policies
The third webinar in the ADS & AMDS Data & COVID-19 series was an international edition with speakers from Penn Wharton Budget Model talking about modelling the health and economic effects of CIVD-19 policies.
The Penn Wharton Budget Model (PWBM) coronavirus simulator projects changes of government policy on business activity and the concomitant virus reproduction number (R) that is a key parameter for epidemiological models. This explicit integration allows policymakers to prospectively forecast the effects of policy changes on economic and health variables. The PWBM coronavirus simulator, thereby, attempts to incorporate the best of detailed practice methods with recent advances in academic modelling.
In contrast, standard practice epidemiological methods are often rich in detail but rely on the “hammer and dance” routine that is adaptive in nature. Under that approach, policymakers myopically reopen the economy in steps and then wait to see the impact on economic and health variables. The hammer-and-dance routine, therefore, is potentially very costly relative to prospective forecasting. In contrast, to more stylized academic models, the PWBM coronavirus simulator is able to incorporate real-time, localized data to explicitly model economics and health variables at the local and national level, of course, depending on actual data availability.
For the US, the model shows that personal (voluntary) social distancing (e.g., wearing masks, meeting outdoors, and plexiglass barriers) is even more important than policy (required) social distancing. In fact, how policy changes are interpreted by the population (i.e., change in beliefs) might be more important than the policy itself (i.e., for a fixed set of beliefs).
The PWBM model could be used outside of the US as well. PWBM is open to working with interested governments and research universities. Interested potential partners should contact PWBM faculty director Kent Smetters at firstname.lastname@example.org.