Skip to content

COVID-19: Ethical Guidelines for the Exit Strategy

Alberto Giubilini

Julian Savulescu

Oxford Uehiro Centre for Practical Ethics

University of Oxford

Supported by the UKRI/AHRC funded project “The Ethical Exit Strategy”

(Grant number AH/V006819/1)

These are the “Main Points” and the Executive Summary of a Statement on key ethical considerations and recommendations for the UK “Exit Strategy”, that is, the strategy informing the series of measures to move the country from the state of lockdown introduced in March 2020 to a ‘new normality’.

The full Statement can be found at

The document has been produced also on the basis of the discussion among academics and stakeholders from different fields (ethics, economics, medicine, paediatrics, mental health, nursing), who participated in an online workshop on the “Ethical Exit Strategy”, held on the 8th of July 2020.








Lockdown is an extreme measure more easily justifiable early on in a pandemic and for a short period of time. At this point, other solutions should be prioritized

Lockdown is burdensome and its public health benefits are unclear. The less restrictive alternatives recommended in this document are likely to strike a more reasonable long-term balance of competing values of lives, health, healthcare delivery, freedoms, fairness, and collective wellbeing. Prioritizing them would be consistent with an ethical principle of selecting the least restrictive option in public health
Shielding the elderly and individuals with certain pre-existing medical conditions


Selective shielding of certain vulnerable groups would infringe on the freedoms of fewer people than full-blown lockdown.   It would have both individual and collective benefits as it would reduce the number of deaths and it would not constitute unfair treatment of these groups. Shielded individuals would gain significant benefits in terms of reduced risk of death.
Implementing more effective testing and contact-tracing, including through contact tracing technology (mobile apps)


These measures have been shown to be effective at containing the virus in countries where they have been successfully implemented. They entail some privacy infringement, but the possible harms are likely to be outweighed by both individual and collective benefits.
Quarantining individuals likely to have been exposed to COVID-19


Quarantine involves no individual benefit for those quarantined and there are some costs for those who are subject to it, but it entails a very large collective benefit. The justification for selective quarantine would be stronger if those subject to it were adequately compensated.
Keeping schools open


Closing schools would unfairly burden children, because they are unlikely to significantly benefit from being confined and the benefit would almost exclusively accrue to vulnerable individuals. They are not major spreaders of the virus, and shielding measures should be used to contain the additional risks they would pose within and outside their households
Adequately protecting and incentivising NHS staff subject to larger risks or burdens


NHS workers are likely to be subject to additional burdens, including additional risk, during periods of increased COVID-19 hospitalization. Fairness requires that risks are minimized through adequate provision of PPE and any significant additional risk is properly compensated when their salaries do not already pay for such additional risks.
Consider the introduction of “immunity passports” Immunity passports would allow liberty-limiting policies to be applied to a smaller number of people by identifying those who are not at risk. The collective benefit would plausibly outweigh the downsides, such as unequal distribution of liberty restrictions
Vaccination policies should maximize the collective benefit of the vaccine, but not necessarily prioritize the most vulnerable Protecting the vulnerable is a priority in vaccine allocation. However, higher need for protection does not necessarily mean stronger claim to access the vaccine first. We need to consider on which groups the vaccine will be more likely to be effective. If indirect protection is more effective, we should opt for that strategy.
Stricter enforcement of behavioural modifications, such as face covering in closed public environments

Face covering is effective at limiting the spread of the virus and entails a very small individual cost. When the individual cost of a measure is very small and the expected benefit is very large, there is a strong ethical case in favour of state coercion




The UK “Exit Strategy”, which aims at safely easing the restrictions introduced in March 2020 to contain the COVID-19 pandemics in the UK, needs to balance different values and priorities, beyond protecting the population from the virus. The task will be made even more difficult by the fact that the Exit Strategy will have to be responsive to likely new spikes of COVID-19 cases, if not by an actual second wave of infections. The response to the first wave has been a strict lockdown involving closure of most business activities, schools, and universities; reduction of services (e.g. public transport, postal services, etc); and requirements to remain at home except for basic needs, and to socially distance from other people whenever outside of one’s home. Both lockdown and the spread of the virus entail large costs in terms of lives lost, health (both physical and mental), and economic damage (job losses, recession, reductions in households’ income). Such costs have been and will likely be unevenly distributed across the population.

Containing the virus with very restrictive measures such as lockdowns is only ethically justified if the benefits outweigh the costs and there are no less restrictive alternatives that could plausibly achieve the same results.

However, costs and benefits ought to be measured not only in terms of health and death toll of COVID-19, but also in terms of overall impact of such measures on the population’s wellbeing. Our ethical analysis is informed by the evidence on the effects of both the COVID-19 pandemics and the response measures adopted so far in the UK and elsewhere. It suggests that while there were strong ethical and public health reasons for imposing a lockdown at the beginning of the pandemic, at this moment there are strong ethical and public health reasons to prioritize less restrictive measure. The measures here listed would strike a more reasonable and sustainable balance among the values of health, life, healthcare delivery, fair distribution of burdens and benefits, freedoms and other individual rights, and ultimately wellbeing at the population level.


Prioritizing alternatives to lockdown: the ethical and public health justification for lockdown at this stage is very weak. Lockdowns entail a very large economic cost, as well as many other types of costs in terms of mental health, educational gaps, and inequalities. Besides, they are only effective at containing the virus if properly implemented and they are more likely to be effective if implemented very early on in outbreaks. They require adequate level of enforcement by authorities and sense of responsibility by individuals. Thus, in some countries that enforced strict lockdowns during the ‘first wave’ – including the UK – COVID-19 mortality rates have been higher than in other countries with much milder restrictions. This might partly be explained by late enforcement of lockdown, or other factors leading to a larger or more deadly outbreak. At this point, there are alternatives to lockdown that can strike a better balance among the different values at stake.


Shielding the most vulnerable individuals. Without lockdowns, restrictive measures should be applied only to certain groups that are more likely to suffer severe consequences from COVID-19. These include the elderly and those with certain pre-existing medical conditions. Shielding the most vulnerable is consistent with a basic ethical principle of least restrictive alternative in public health: under a plausible understanding of this principle, we ought to adopt the policy that can satisfactorily limit the damage of COVID-19 and that is the least infringing upon individual freedoms – including infringing upon the freedoms of the lowest number of people possible. It is not unfair if the criterion for shielding is need for protection and individual benefit to the shielded.


Implementing effective testing and contact-tracing, if necessary through mandatory use of contact-tracing technology. Contact-tracing is one of the most effective ways of reducing deaths and hospitalizations caused by COVID-19, especially if implemented together with the kinds of interventions we are suggesting here. Technological solutions like contact tracing apps, if properly implemented, are very likely to be effective. They entail some risk of privacy infringements. Privacy is an important value in our society. However, returning to some form of normality where privacy protection can be restored to pre-pandemic level might justifiably require making sacrifices in terms of privacy in the short term.


Quarantining and isolating individuals reasonably presumed to have been exposed to COVID-19. Confining individuals who are more likely to be infectious is another essential measure to contain the spread of the virus that applies the least level of restriction possible at the population level. Such measures should target people who test positive to COVID-19, people who have been tracked through tracing procedures, travellers returning from areas with high infection rates. If quarantine is implemented selectively, there is a strong ethical case to compensate those who are subject to it.


Keeping schools open. Children are among the individuals who have more to lose and less to gain from school closure, since they are extremely unlikely to suffer major consequences of COVID-19 and would pay a high cost in terms of missed educational opportunities. School closure, after an initial period when it might have been justified, would unfairly burden them. Shielding the vulnerable people whom they might otherwise infect would better protect their interest while reducing the public health impact of their possible exposure to COVID-19.


Better enforcement of face-covering requirements. Using face masks is not as culturally accepted in the UK as it is elsewhere, e.g. in Japan. However, it is an effective way of limiting the spread of the virus – actually, the successful management of the pandemic in Japan can be explained also by the traditional widespread use by Japanese people of face masks to prevent spread of diseases. Since the cost to individuals is small and the collective benefit is large, there is a strong ethical case for using state coercion in enforcing face-covering requirements.


Better supporting NHS staff exposed to higher risk and to increased working hours. Throughout this pandemic, NHS staff will at times need to take on additional burdens, in terms of increased workload and/or additional risks due to more likely exposure to COVID-19. Providing them with adequate Personal Protective Equipment is essential, but fairness requires that any additional risks and workload be properly acknowledged and remunerated when such risks and burdens are not already reflected in their salaries. The best way to guarantee fair work arrangements in this situation is to have an incentive scheme for NHS workers who voluntarily take on additional risks or workload.


Immunity passports ought to be considered Immunity passports would allow immune people to engage in certain activities from which the non-immune are excluded. Once again, this solution would limit the liberties of a smaller number of people than other liberty restricting policies. The collective benefit both in health and economic terms would plausibly outweigh the downsides of immunity passports, such as the unequal distribution of liberty restrictions and some privacy infringement. When we have a vaccine that is available to everyone, the ethical case for immunity passport will be stronger because immunity could be acquired in a relatively safe way. Confining everyone when we could safely confine a smaller number of people would be a form of ‘levelling down’ equality, which is ethically impermissible in this case.


Vaccination policies will need to take into account how effective the vaccine will be on different groups. It is reasonable to say that to maximize the benefits of the future COVID-19 vaccine, we will need to distribute it in a way that protects in the first instance the most vulnerable. However, this does not necessarily mean that the most vulnerable ought to be prioritized in accessing the vaccine, while availability is limited. A lot will depend on how well the vaccine will work on different groups. Vaccines can protect individuals directly or indirectly (through herd immunity), and we will have to figure out which of the two is more likely to protect the vulnerable, given initial limited availability.



The full document is available here.




Share on

1 Comment on this post

  1. This document on first blush makes some assumptions that all of the population are geographically stable and will remain that way, choosing decision criteria most appropriate to that assumption. In the same way the presented outcome appears to assume that ethics should most appropriately reflect a given and focused morality. Such outcomes are heavy influenced by political decisions, rather than morally justifiable conclusions for humanity as a whole. Whilst it is accepted that existing national frameworks form an underlying foundation; and policing measures could be used to reduce population movement; and that a level of prior notice does allow those influencers most able to remove themselves from higher risk areas; those actions do not appear as ethically fair and so have been ignored for the purpose of the above comment.

    Example offering illustrations:
    If the pandemic continues through the winter, as appears to be being accepted, then a logical extension of the current COVID-19 rules would result in the larger UK population centers all being ‘locked down’ (a truly awful term) and the rural areas remaining more open, which could lead to overwhelming of the health services in rural areas. (Something the local politicians within Nottinghamshire appear to have recognized as they state rural areas should quarantine in the same way that Nottingham City is to be.) One only has to consider the conduct of high profile cases (Mr. Johnson – Chequers; Mr. Cummins – Barnard Castle?) when faced with decisions regarding the life, health and well being of their own families, once the public effort was temporarily removed from them, to see that element of simplifying the risks to life at play.

    Whilst the spread sheet generation often reflects, and remains reflected in historical data, it appears sensible to look forwards towards personal difficulties others will experience in the variety of their own ethical choices, especially when those most able and informed provide such visibly contrary examples. A self limiting focus based upon organisational needs set within national limits does not achieve that, but may answer short term political demands. Is it possible for ethics to become more than a resource availability and control system?

Comments are closed.