Is this proof that social distancing makes things worse?

Increasingly it is becoming clear that far more people are resistant to the Wuhan coronavirus than was initially assumed, and thus that collective immunity is achieved at a much lower infection rate than 60 per cent.

Ross Clark in the Spectator reports on a study in the journal Cell which finds 40 to 60 per cent of a control group not exposed to Covid-19 nonetheless having T cells (which play a central role in the immune response) that give them ‘some degree of crossreactive, pre-existing immunity’ to the virus.

This backs up findings from Germany that 34 per cent of the population have T cell cross-immunity from other viruses, and a study from Singapore that finds nine out of 18 people showing a T cell response to a Covid-19 virus protein.

Some scientists are now suggesting that this explains why some countries, such as Germany where Covid deaths per million are less than a fifth of the UK’s, get off more lightly than others. Karl Friston, a neuroscientist who is connected to the independent SAGE (not the official SAGE), told the Guardian:

‘There are various possible explanations, but one that looks increasingly likely is that Germany has more immunological “dark matter” – people who are impervious to infection, perhaps because they are geographically isolated or have some kind of natural resistance.’

This is important for understanding the behaviour of the virus, as it is becoming increasingly clear that social distancing measures explain little if any of the difference in outcomes between countries, and thus that in most places the virus is running its course and dying down, not because of interventions but because it reaches a saturation point or collective immunity threshold much earlier than was expected.

We can see the evidence for ourselves. If social distancing did the trick we would expect countries which adopted it earlier to do better overall. We’d expect the curve to flatten earlier and we’d expect a lower death toll per million of the population.

In the two graphs below I’ve plotted these numbers for seven countries and one US state. On the X axis along the bottom is a measure of how early social distancing was adopted in a country, calculated by finding how many days elapsed between the first Covid-19 death in the country (when the 7-day rolling average reached 1 death) and when public transport use fell to 50 per cent of its pre-epidemic level. On the Y axis up the side is, in the first graph, how quickly the curve began to flatten, calculated by finding how many days elapsed between the first death in the country and when the daily growth rate began to decay from an exponential increase. On the second graph the Y axis is the Covid-19 deaths per million.

Source: Government websites and Worldometer (death data); Moovit and Citymapper (public transport data)

If social distancing made things better, we would expect a positive correlation on both of these graphs – in other words, earlier social distancing would lead to both earlier flattening of the curve and lower total deaths, meaning these points would all sit close to a diagonal line sloping up from left to right. Instead what we see is very little correlation at all, and what there is is negative. So early social distancing is either doing nothing or making things worse. This is likely because the virus spreads mainly in hospitals, care homes and private homes rather than in the community, so social distancing of the wider population beyond a basic minimum (washing hands, self-isolating when ill, not getting too close, and so on) has little impact. The countries with the highest death tolls are often those which fail to protect their care homes adequately, with up to 82 per cent of Covid-19 deaths occurring among care home residents.

In the graph below, however, we do find a positive correlation. This plots the population size of the country on the X axis and, again, the time between the first death and the flattening of the curve on the Y axis.

Source: Government websites and Worldometer

The correlation here suggests that the flattening occurs later in larger countries and earlier in smaller countries, which is what you would expect if it was occurring largely because it was approaching a natural saturation point or collective immunity threshold (since larger countries take longer to reach the threshold). This backs up natural rather than social causation.

With the UK looking likely to be among the worst affected countries in the world, despite government predictions that our early and tight lockdown would get us off lightly, it needs to be asked what our stupendously costly lockdown has achieved, and what is being gained by emerging so slowly.

It is tempting to say that politics long ago took over the driving seat from science in the government’s handling of this crisis. But that would be to ignore the continued calls from government scientists to go slowly and be cautious. The truth is that the country has been badly let down by its scientific advisers throughout this pandemic, most of whom seem to cling to discredited models and be impervious to new evidence and data, showing themselves all too willing to wield the precautionary principle as a tyrant. Something has to change.

First published on Conservative Woman.

2 thoughts on “Is this proof that social distancing makes things worse?

Add yours

  1. Thanks for this Will. I think some of your stats are interesting…though I would want to ask why you have selected just these examples to compare? What happens when you add others in?

    And how do you account for David Spiegelhalter’s statistical analysis, which showed that expected deaths from Covid-19 matched exactly expected deaths by age for the coming year?

    If he is right (and it is a good analysis) then we would have seen 600,000 deaths in the space of a month. yet that didn’t happen, and that must have been because of social distancing in some way or another.

    again, there were clear mass outbreaks linked with big gatherings, so if these had not been banned, the disease would have taken a far bigger toll.


  2. ” right (and it is a good analysis) then we would have seen 600,000 deaths in the space of a month. yet that didn’t happen, and that must have been because of social distancing in some way or another”

    His analysis is about the risks to a person if they get COVID, and insofar as that is a very good analysis. It doesn’t make claims about how many people might get COVID, as far as I know, or if it does it assumes a naive herd immunity of 60 percent.

    This assumption is probably incorrect; it would’ve never reached 60 percent of the country’s population, for a number of reasons. Firstly the population isn’t homogenous in their spreading of the disease, 10 percent cause 80 percent of cases in clusters. Once those 10 percent have had it and are immune, the spread of the disease is rapidly limited.

    The second reason is that the population is not homogenous in their susceptibility to the disease, some likely have prior immunity from genetics or from other coronaviruses. This might reduce the population who can get the disease by 50 percent.

    The spread of the disease was never exponential, the exponent declines rapidly from before social distancing came into effect. I don’t know what caused that; I’m not qualified to suggest a reason, but it’s not clear that any public health measures had much effect, although they may have. What effect they had didn’t appear in the change of rate of increase of this disease. Where else they might show an effect I do not know.

    What is not deniable is that the 600000 deaths was a wild overestimate; we haven’t had that many in the world yet! It’s out by orders of magnitude, most likely. Models have shown to fail the test of prophecy.



Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Website Powered by

Up ↑