Science and policy in extremis: the UK's initial response to COVID-19

Abstract

Drawing on the SAGE minutes and other documents, I consider the wider lessons for norms of scientific advising that can be learned from the UK's initial response to coronavirus in the period January-March 2020, when an initial strategy that planned to avoid total suppression of transmission was abruptly replaced by an aggressive suppression strategy. I introduce a distinction between "normatively light advice", in which no specific policy option is recommended, and "normatively heavy advice" that does make an explicit recommendation. I argue that, although scientific advisers should avoid normatively heavy advice in normal times in order to facilitate democratic accountability, this norm can be permissibly overridden in situations of grave emergency. SAGE's major mistake in early 2020 was not that of endorsing a particular strategy, nor that of being insufficiently precautionary, but that of relying too heavily on a specific set of "reasonable worst-case" planning assumptions. I formulate some proposals that assign a more circumscribed role to "worst-case" thinking in emergency planning. In an epilogue, I consider what the implications of my proposals would have been for the UK's response to the "second wave" of late 2020.

Keywords: COVID-19; Science and policy; Science and values; Scientific advice.

Fig. 1

A sketch of a graph from SAGE (2020d), dated 9 March 2020. The “high transmission reduction” line depicts a strategy in which herd immunity is achieved by the summer through natural infection. The “very high transmission reduction” line depicts a strategy in which aggressive suppression in the spring leads to a disastrous wave of infection in the autumn, again leading to herd immunity through natural infection

Fig. 2

Projected demand for critical care and critical care capacity, from Ferguson et al. (2020). The projections indicate that, without aggressive suppression measures such as school closures, demand for critical care would exceed supply many times over

Fig. 3

A graph from Ferguson et al. (2020), in which the blue-outlined blocks represent periods of school closure. This is a projection of the consequences of closing schools for about two thirds of the time, indefinitely, until a vaccine is developed