The world still has a way to go with responding to the global covid-19 pandemic, but can we discern factors that have determined how well different countries have done in responding so far? Ngaire Woods, founding dean of the Blavatnik School of Government and professor of Global Economic Governance at the University of Oxford, recently gave a talk to the Royal College of Physicians of Edinburgh in which she attempted to answer the question. Her answers made sense to me and will, I think, to others.

She began her talk by showing the ineffectiveness of three attempts at measuring pandemic preparedness. High income countries generally outscored low-and-middle-income countries, and the UK and the US came top in one of the measures. Yet these two countries did poorly with their responses in the first year and have been “saved” (so far) only by the vaccination programmes.

Woods identified three governments from low-and-middle-income countries that have done particularly well: Senegal, Vietnam, and Sri Lanka. They all managed to move fast with introducing track, trace, quarantine, and support programmes. Woods emphasised that all four are needed for effectiveness. They have different political systems, but they have all had experience of recent pandemics.

Assessing the performance of countries has, observed Woods, given everybody a chance to advance their pet theories. Have authoritarian governments or democracies done better? Have countries with women leaders done better? Woods sees nothing convincing at this level, and even the extent of universal health coverage, a popular theory, does not seem to explain responses: Iran, for example, has close to universal health coverage, but has done badly.

Woods identified three factors that had a strong influence on how well countries did. Firstly, the presence of “warring elites” is associated with poor outcomes. There were “warring elites” and turf battles within national governments, between subnational and national governments, and among academics, scientists, epidemiologists, and doctors. Academics like me, she said, are trained to argue with each other. She cited the arguments over face masks, suggesting that perhaps less attention should have been paid to the need for randomised trials and more to the empirical evidence that countries that had experienced pandemics promptly encouraged the use of face masks. (She didn’t mention it in her talk, but Oxford academics were leaders of “warring elites.”)

The single most important factor, she argued, was effective collaboration between national and subnational governments. (Many of those listening to the talk must at this point have conjured up pictures of Boris Johnson arguing with Andy Burnham and the devolved nations moving in different directions.) Effective collaboration depended on having effective coordination bodies, cooperation among regions, continuous dialogue, transfer of resources from national to subnational government, and cooperation in recovery plans. The US, UK, and France all did badly with collaboration between national and subnational governments, while Germany did better.

The third factor important factor was collaborative leadership, recognising that “we can’t do this alone,” and inviting politicians, frontline health workers, epidemiologist, police, everybody to contribute. Woods picked out Greece as having done well with collaborative leadership. Such leadership was lacking in the UK.

Woods was perhaps bold in accepting the Edinburgh college’s invitation to reflect on governance (“proper function of institutions and their acceptance by the public”) in the covid-19 pandemic, but her observations chimed with me and many others. I recommend that you watch the talk.

Richard Smith was the editor of The BMJ until 2004.

Competing interest: RS is a fellow of the Royal College of Physicians of Edinburgh and about to discuss governance and climate change as part of the College’s programme on governance.