Should we have another tough lockdown? Not yet

Starting a lockdown too early or too late reduces the benefits but not the hardships

By Alex Welte

31 March 2021

The streets of Cape Town were eerily quiet on the first day of the March 2020 lockdown. Archive photo: Ashraf Hendricks

President Ramaphosa introduced comparatively minor lockdown restrictions on Tuesday night: limits on the size of gatherings, and no alcohol off-sales over the Easter weekend.

So when does it make sense to introduce tougher measures?

As of 31 March, infection rates are bumbling along with minor daily fluctuations – showing signs neither of fading further, nor of resurging. Unless we think we are about to hit some sort of jackpot, infection rates can only go up. This means we have some decisions to make so that if the next wave turns out to be another big one, we don’t miss a chance to take the edge off.

One year on, there is still heated debate about lockdowns, and various vested interests are understandably gearing up to fight any whiff of further tough measures against this or that particular industry – catering, alcohol, travel, tobacco.

But we know that some lockdown measures achieve their intended aim of making it difficult to transmit SARS-CoV-2 – the virus that causes Covid-19. We also know that lockdowns bring ‘collateral damage’ in the form of jobs lost, contracts not wanted, companies making losses and going under – and the way these costs are borne/distributed among the population is partly a political matter, not just some sort of revenge of the fates.

So the question becomes – is there a way to use lockdown measures carefully, so that the direct and indirect costs are bearable? And to this there is no uncontroversial answer – because we cannot objectively attribute values to all the relevant ‘costs’, ‘effects’ and ‘benefits’, as they are called by health economists.

But we can at least see that some options are clearly better than others.

Some have claimed that lockdowns merely delay, but do not reduce (in the long run) infections, and so really just buy time to get the health system ready to deal with the peak of the coming wave. This is partly true, but largely false.

If an early lockdown is just an isolated temporary measure, accompanied by no other precautions like reducing work contacts, wearing masks, reducing social gatherings, then, indeed, it has very little effect in reducing the total number of infections that accumulate in the long run. Why? Because, at the end of lockdown, the system will look very much like it did at the beginning, or, at best a little before the beginning of lockdown, and from this ‘beginning’ the trajectory just resumes.

If, however, we institute fairly strict lockdown measures really close to the peak of an epidemic, then there can be a much bigger impact - not just during the period of restrictions, but in terms of overall number of infections accumulated in the long run.

Here is a quick recap of what has been explained numerous times this past year:

Epidemic peaks are intertwined with immunity. If there is no immunity, people will just keep getting reinfected. But this is not what is happening with Covid. Most people, once infected, have some immunity to reinfection, at least for some time.

As people acquire immunity, the remaining infected individuals are less and less likely to come into contact with non-immune people to infect, and so outbreaks, after reaching sharp peaks, fade.

Here’s another way of putting it. Let’s say on average that infected people make sufficient contact with two people, over the course of their own infection, to pass on the virus. This means the reproductive number, R, is 2.

Now what if, at some point in time, more than half the population has somehow acquired immunity? From that time on, infected individuals will mostly make contact with immune people and therefore fail, on average, to infect more than one other person over the course of their own infection. This means the reproductive number, R, has fallen below 1. In this scenario, the epidemic fades away.

We can now analyse how lockdown measures can be used effectively. Imagine we’ve taken a deep breath and avoided serious restrictions until the rate of infections is alarmingly high. Now we institute a sudden lockdown and interrupt the activities which bring people together and make transmission possible. During this interruption, we don’t fully wipe out all infections, but many, maybe most, of the people who were infected do in fact recover, and become immune.

We will have achieved two things: 1) dramatically reduced the number of infectious people, and 2) substantially increased the number of immune people. If our timing is ‘perfect’ we will emerge from lockdown with enough collective immunity to have pushed that reproductive number below 1. Compared to what would have happened without any lockdown, the ride down the back of the peak will have begun at a smaller cumulative count of infections, and will add less to the total.

One can be forgiven for thinking that perhaps the timing is ‘critical’ – in the sense that we have to somehow ‘get it just right’ to get the benefit – but it’s not like that. The models show that even an early lockdown has a bit of an effect in reducing the ultimate number of infections. Then, as the timing approaches the ‘sweet spot’, the impact increases. Even if the lockdown comes very late in the epidemic, there is some net reduction in the number of infections.

Can we quantify all this reliably, and propose some sort of perfect-timing formula? Alas, no. Not only do we not know enough about an unfolding epidemic – we don’t even agree on what is an acceptable cost, or a valued effect.

What is clear though, and what everyone should be able to agree on, is that a tough intervention when a wave has not even perceptibly begun, or a tough intervention when it is essentially over, does not buy us much. It will be better to hold our nerve, and make the sacrifices, when, indeed if, the next wave gets really bad.

Welte is Research Professor at, and the former Director of, the South African (National Government) Department of Science and Innovation - National Research Foundation (DSI-NRF) Centre of Excellence for Epidemiological Modelling and Analysis (aka SACEMA), at Stellenbosch University.