Everyone wants to know when Covid-19 will “end”. The short answer is when we find a vaccine and administer it to 80-95 per cent of the world’s population, which will be mid-2021 at the earliest. Twelve months is a long time to be making crucial Covid-19 decisions in a volatile and unpredictable environment. Lockdowns have generally worked, but they are blunt instruments and cannot last till mid-2021. We all want to restart our economies and societies as safely, quickly and widely as possible. How should governments decide how, how much, how fast and which sectors to restart? I offer three suggestions. If implemented together, we can create a steady state of predictable decision-making until we deliver a vaccine to all 7.7 billion human beings. The first suggestion is to be more predictable in the time schedule of decisions. Right now, the general trend is to make reactive decisions to lift lockdowns or restart economies based on ad hoc events, whenever economic or political pressures become overwhelming, or when epidemiological indicators show the smallest sign of recovery. These unscheduled decisions make it difficult for businesses to plan ahead and for citizens to adjust psychologically. Such decisions hold public health policies hostage to the whims and fancies of politicians and commercial interests, while ignoring data. Unscheduled decisions may have been unavoidable in the initial stages of this pandemic but are becoming less defensible. As we enter the fifth month of Covid-19 with at least another 12 months to go, we must move towards more predictable decision-making time schedules. These should follow a set rhythm, for example every Monday, every second Monday or every first Monday of the month. The predictable timing of decisions has many advantages, with the main one being enforced gradualness in the national restart. It will also reduce the risk of decisions being captured by lobby groups, increase the transparency of decision-making and reduce decision fatigue by governments, not to mention depoliticising decisions. The second suggestion is to be more predictable in the criteria used to make decisions. Right now, decision makers are using ever-changing or cherry-picked criteria to support their own biases. So, instead of evidence-based policymaking, we have policy-based evidence making (this is perhaps the polite term for leaders lying using manufactured statistics). Inconsistent criteria could have been unavoidable at first, given the bureaucratic silos of any government, but this has also become less defensible as time goes on. A consistent set of criteria until mid-2021 will introduce more predictability for citizens and businesses, although it will be politically contentious to create this set of criteria. Coronavirus: are travel bans an effective way to curb Covid-19’s spread? There are two broad solutions to find that balance between lives and livelihoods. One is to depoliticise the criteria by relying on hard epidemiological data such as the number of new cases, number of active cases or incidence ratio in the last 14 days. This must be backed up by expanded testing programmes to improve the accuracy and reduce the time lag of the data. Another is to schedule a periodic review for the relevance, usefulness and timeliness of the criteria, for example every three months. This will hopefully defuse the politics just enough to write the predictable set of criteria into the long-term national strategy. The biggest benefit of using predictable epidemiological criteria is that decisions are more immune to economic or political pressures. The third suggestion is to reopen by sectors, not by geographies, which will resolve the challenges of geographic reopening. For example, the national-level strategies of Brazil, the Philippines, the United States, and Malaysia are not supported by all state or provincial governments. A geographic reopening must resolve the endless debate of the right size of the “administrative and implementation unit”. Should it be restarting the whole country, or by states, regions, districts, cities, towns, or villages; and who has the authority to make the decisions? Each of these administrative units have different governments, resources, enforcement abilities, risk profiles, disease burdens, and economic activities. Hence, a national-level restart could be dangerous, unfair, and unpredictable. The West could have made Covid-19 plans in January. Why blame others now? A sectoral restart is an elegant solution to the complexities of federal-state relationships. For example, governments can jointly decide on essential, semi-essential, and non-essential economic activities, and restart the economy in phases based on this. Not only that, governments can restart the economic sector first, while announcing that social activities (funerals, religious congregations or restaurants) may restart in one month, followed by the educational sector (starting with children in examination years) in two months, the cultural sector in three months, and so on. This is just a rough guide. The main point is to reopen sectorally, and announce the sequence so that citizens and businesses can prepare for it. These three suggestions are interdependent and should be implemented together. However, they do not resolve other problems such as the quality of political leadership, available political will, government transparency and accountability, the resilience of the health system, and the quality of decision-making criteria. These require other solutions. Coronavirus: why Asia will win the race to economic recovery What’s clear is that after five months of fighting Covid-19, we need to rapidly move into a predictable, steady state of decision-making. Reactive policies on the back of decision fatigue are real problems, not to mention being ineffective and dangerous. Predictable timing of decisions, criteria for decisions and reopening by sectors will help us achieve that steady state of decision-making. Help us understand what you are interested in so that we can improve SCMP and provide a better experience for you. We would like to invite you to take this five-minute survey on how you engage with SCMP and the news.