Do lockdowns hinder or help? Detailed analysis shows there is NO excuse for inflicting them on the People

A THOROUGH, SCIENTIFIC AND “NERD FRIENDLY”ANSWER FOR YOU

Introduction

by Fabian

Having asked it in a previous article, we ask the question again: do lockdowns work?

Here’s a thorough, scientific and “nerd friendly”answer for you.

SOURCE

Do lockdowns help?

No, they only appear to make things worse

Frans Rautenbach has analysed whether lockdowns help with preventing the spread of Covid-19, and has come to the conclusion that they don’t.
To find out whether lockdowns work to combat the Covid-19 virus, one can calculate the strictness of the lockdown measures of countries, and plot those results against the number of deaths from the disease per million people per country.

Our World in Data has an Index that measures the stringency of government responses to the pandemic (the Stringency Index for short) that is based on, “… nine response indicators including school closures, workplace closures, and travel bans, rescaled to a value from 0 to 100 (100 = strictest)”.

graph 1

The period of measurement runs from 21 February 2020 to 21 September 2020, the end, more or less, of the first wave of the pandemic, when in most countries the number of deaths per day approached nil. The stringency number is worked out by calculating the average weekly figure over 31 weeks.

graphic 2

graphic 2

From the above comparisons it is clear, at the very least, that there is no correlation between the stringency of government interventions such as lockdowns, and low death tolls from Covid-19.

On the contrary, the correlation runs in the opposite direction. Lenient lockdowns are associated with low death totals.

However, one remains mindful of the principle that correlation does not prove causation. So, lockdowns don’t necessarily cause high death rates. It may be, for example, that stringency of lock-down-type measures was caused by the perceived threat posed by rising corona death figures, rather than the other way around. Governments conceivably imposed stricter measures precisely because they were under pressure due to observed rising deaths. I return to that possibility below.

At this stage, what the table shows is that lockdowns at the very least appear not to be a key determinant of the spread and resultant mortality caused by the disease.

That question is returned to below too.

One argument that is bound to crop up at this stage of the discussion, is that lockdowns have justifiably been strict at the outset of pandemic outbreaks, in that they were necessary to curb the onset of cases so that hospitals and other health facilities could cope.

Timing of lockdown

In other words, the average strictness of a lockdown is less important than the timing of its strict application. The narrative around New Zealand, for example, seems to support this notion. New Zealand had a strict lockdown up front (soon to be relaxed), and as everyone knows, very low death statistics.

One has to ask, however: If “flattening the curve”, at the outset was such an effective measure, why did the countries with high death tolls not demonstrate that?  Almost all the countries in the table with high stringency measures (the first and second quartiles in particular) imposed strict lockdowns within a week or two of the start of the pandemic in those territories, and all about a week before their daily cases peaked, and virtually all have high death percentages. What is more, many island states in the Pacific Ocean like New Zealand did not follow such an approach of severe lockdown up front, followed by a relaxation thereafter – and they ended up with even lower mortality rates by 21 September, such as Singapore, Fiji, Papua New Guinea, and Taiwan.

That prompts us to say that the acid test is not countries that imposed strict lockdowns from early on, irrespective of whether they lifted those measures early.

The question is rather how countries fared that never imposed strict lockdowns, whether in the beginning phases of the pandemic, or at any other stage. Based on the prevailing narrative, those countries would have unleashed the fury of the pandemic on their unsuspecting populations, causing havoc.

It turns out there are six countries in the world whose response stringency never exceeded 50 on the Stringency Index. In other words, they are the six countries that we know about, with the most relaxed government responses of all countries in the world. Their maximum stringency measures of 50 or less, compared to the typical 80 in the above sample of 24 diverse countries. Even controversial Sweden started with a response of 64, significantly higher than the 50 used as my cut-off point in the sample of six. These countries’ maximum stringency figures are shown below, together with their confirmed deaths at the end of the first wave at 21 September 2020. As a control, I compared the figures with those of three neighbouring peers of each.

graph 3
graph 4

As can be seen, the average of these countries’ deaths from Coronavirus as at 21 September 2020 was 15 per million people. That can be compared to the world average at that stage of 123 per million people, the average of the peer countries of 31, not to mention the typical locked-down, developed country which had hundreds of deaths per million at that stage (such as Belgium with more than 800).

The obvious significance of these comparisons is that it shows that those countries that had the mildest of possible lockdowns and similar measures at the outset of the pandemic (and at all times thereafter), had such low death percentages that they were almost negligible.

Stringency 

Now the ball game changes. We have to ask why countries with the most modest of lockdown measures, namely the six least locked-down societies in the entire world, not only did not have worse outcomes, but outperformed their stricter counterparts – in most cases dramatically? Can we still argue that high or rising mortality explains strict lockdowns, not the converse?

The figures do not bear this out. In the next table I track the gradient of increases of new daily cases in the last week preceding the imposition of the most stringent measures, in each of the 24 diverse countries above. The next two columns then respectively show the average stringency over the period, and the maximum stringency imposed by each country.

graph 5

If the explanation for the high-lockdown/high death total correlation was government concern about rising infections or death tolls, this table would confirm it. Yet there is clearly no such correlation. In other words, severe lockdowns were not caused by governments spooked by rapid growth in infections and resulting deaths.

The same exercise can be done in respect of the six low-lockdown countries and their peers.

graph

graph

Again, there is no correlation between preceding high rates of case growth, and strict lockdowns. On the contrary, the low-lockdown states on the whole had higher prior growth than the high-lockdown peers in the sample. If the theory of strict lockdowns following scary rises in cases had any validity, it should be the other way around.

To test the same theory, we can also consider a random sample of countries in the low-lockdown group and that some peers bear – but based on preceding death rates this time. For example, on 4 April 2020 Russia had fewer deaths per million than Belarus (.29 vs .56), but imposed about seven times stricter measures (87 vs 12), ending up on 21 September 2020 with 150% as many dead.

As at 31 March both Tanzania and Kenya stood at 0.02 deaths per million. Yet at that stage Kenya’s stringency was almost double that of Tanzania (87 vs 47), ending up with a death total about 30 times higher. When China commenced imposing measures at about 26 January 2020, its death toll stood at 0.05 per million, barely worth working up a sweat about. Yet when Taiwan stood at the same death rate (about 20 March) it had a stringency of less than half of that of China at a comparable stage (31 vs 67). China ended up on 21 September with ten times as many per million dead as Taiwan, and many more than the nil recorded by Macau.

In Italy, poster child of failed responses to the pandemic, when the death rate was less than one per million on 1 March, its stringency figure already sat at 69. On 29 March 2020, when South Africa’s daily mortality was less than 1 per million, its stringency was already a severe 87. It remained over 80 until 17 August, during which period total deaths increased exponentially to over 200 per million.

These examples show that the correlation between lockdown stringency and mortality is on balance not explained by governments responding more severely due to higher death rates. Many countries imposed strict lockdowns long before mortality was high, and those who did not, fared better on the whole.

Of course, there are some countries where governments imposed strict lockdowns in response to observing exponential mortality curves, albeit from a low base. Examples of these are the UK and Spain. But at best that proves that those governments were spooked by rising figures, prompting them to impose severe lockdowns. It tells us nothing about whether their death rates would have been better or worse without those strict measures. The only way to answer that question, is to observe countries that in fact imposed next to no lockdowns, like the six in the sample above.

Is there a rational, scientific explanation for their low mortality? Why would countries with far more relaxed lockdown-type measures have lower mortality than comparable countries with much stricter measures? Could it be that it was the very imposition of lockdowns that was causing the higher death rates? And if so, why?

Climate not a roleplayer

What we should say immediately, is that the list of six contains a 50/50 split of diverse countries (Tanzania, Burundi, and Belarus) and homogeneous countries (Japan, Taiwan, Macau). So, homogeneity is not the reason for the low mortality here.

Nor is climate. While Tanzania, Burundi, and Taiwan are hot, Belarus and Japan have very cold regions.

Nor is low population density the key. Japan, Taiwan, and Macau have very highly dense urban areas, yet did not seem to require strict lockdowns.

Finally, by comparing the six low-lockdown countries with local peers, any suggestion that regional location is decisive, is refuted. Although it is true that Africa and the Far East have low mortality figures overall, the low-lockdown countries in Africa and the Far East even outperformed their peers in those respective regions.

That leaves at least three possible explanations worth considering:

  • Lockdowns create mini “super-spreader” events by confining more people indoors, or forcing them to converge in large numbers at government offices, border posts, and airports;
  • More lenient measures result in herd immunity being acquired more quickly; and
  • Populations with fewer lockdown measures behave more sensibly, because they have no false sense of security induced by government interventions, and so accept more personal responsibility for their own health.

As a libertarian, my instincts plump for the third explanation. When left to their own devices, people take responsibility for their own (and others’) lives. When nannied by the state, they develop a misplaced belief that the government will protect them – or worse, they rebel against lockdowns or similar regimes and undermine them.

Ultimately, the answer is beyond the scope of this article, and the expertise of this author. But what is clear, is that:

  • There is little evidence that lockdowns help;
  • On the contrary, the liberal approach of countries like Tanzania, Japan, Belarus, Burundi, Macau ,and Taiwan, for whatever reason/s, somehow resulted in far lower mortality rates from Covid-19.

– Frans Rautenbach is an advocate and author of ‘South Africa can work’. He writes in his personal capacity.

Do lockdowns hinder or help? Detailed analysis shows there is NO excuse for inflicting them on the People

Facebooktwitterredditpinterestlinkedinmail