News in Charts: Spread of COVID-19 is slowing for reasons other than lockdown

The economic recovery henceforth depends largely on the ability of governments to suppress the spread of the virus. Countries such as Germany and South Korea – that dealt effectively with the virus – are now able to focus on nursing their economies back to health, whereas states in the US that rushed to lift lockdowns have been forced to re-impose some restrictions as the spread picks up again.  We therefore felt it necessary to try and understand the factors that influence ‘R’ – the rate of infection – and estimate a statistical model in order to do so.

We find strong statistical evidence that the rate of spread of COVID-19 in countries affected by the virus has fallen, on average over time, for reasons unrelated to periods of lockdown. Crucially, the spread of the disease appears to play a role too. We find that as the number of deaths per capita rises in a country, the multiplier falls.

The chart below uses the model to show how the reproductive rate might have evolved in the UK (i) because of reduced mobility alone and (ii) because of reduced mobility and the fact that transmission tends to reduce over time as the number of cases rises. This chart is not drawn using UK-specific data on transmission, but rather by assuming that the UK behaves just like the average of all countries. It suggests that, were it not for this unknown factor, which we are proxying by the cumulative number of deaths as a proportion of the population, it is likely that the increase in mobility over recent weeks would have pushed R above 1 in the UK. While the UK has seen spikes in a few places, with the city of Leicester recently going back into lockdown, it is nevertheless true that recorded infections continue to fall at the national level, albeit slowly.

IN-HOUSE CHART

Recent developments in the US, where the number of daily new cases has begun to accelerate after plateauing for several weeks, arguably poses a challenge to this finding. To investigate further we re-estimated our global model using only data from the 50 US states, and from the District of Columbia. There were small changes in some of the coefficients, including a slightly lower starting value for R. But our main finding, that the rate at which the disease spreads falls over time as the proportion of the population already infected rises, still holds. As our second chart shows, those states that have seen the biggest acceleration in new cases in recent weeks are, by and large, also those that had seen fewer cases to date. This is not a second wave in the US, but a rippling out of the first one.

IN-HOUSE CHART

The statistical result that we have found should come as no great surprise. It simply captures, in a formal sense, what we can observe in the data. Following the gradual easing of lockdowns in those countries most seriously affected by COVID-19 to date, there has been no sustained surge in infections as our final chart makes clear. This gives us some hope that the global economy can continue its cautious reopening, and steadily return to more normal levels of activity. That is why, over the past few months, we have increased the weight we attach to a so-called ‘V-shaped’ recovery from 40% to 60%.

Refresh this chart in your browser | Edit the chart in Datastream

Want more charts and analysis? Access a pre-built library of charts built by Fathom Consulting via Datastream Chartbook in Eikon.

Join us for our next partnership webinar with Fathom and Refinitiv, where we will outline our views on the recovery that lies ahead this Friday 10 July 2020, 4:00pm GMT: Fathom’s Recovery Watch: biggest global recession to be followed by sharp bounce back?

__________________________________________________________________________________