Devise policies to deal with the medical issue and you might have negative economic externalities which could include mass unemployment, loss of livelihoods and increased default skewed towards the more economically vulnerable areas of society; low skilled workers, those with little savings and small businesses. Fail to deal with the medical issue and you have negative medical externalities, which might include increased mortality skewed towards the elderly and the risk of overburdening the health service no longer able to deal with every day medical problems.
Based on the evidence that they have been presented with thus far, Governments around the world have decided to take strong action to deal with the medical issue and attempted to mitigate those negative economic externalities with huge amounts of debt-financed fiscal spending – a policy that comes with future negative economic externalities in itself.
Now, you wouldn’t take medical advice from an economist and nor would you take economic advice from your doctor. Being neither of those things, I am grossly under qualified to pass opinion however, in my role as a fund manager, I am fortunate enough to speak daily with people who are experts in both. Critically analysing this information is what allows me to develop my investment strategy and being party to such information is what allows me to write a piece like this.
The reason I earlier write ‘based on the evidence presented thus far’ is because this is a very important point and so far the data being used to answer the questions ‘how fast will this spread?’ and ‘what is the mortality rate?’ is scant.
Firstly taking the question of how fast will this spread; as part of all of us becoming experts in epidemiology we have learnt the meaning of the phrase R0, the basic reproduction number – this is the average number of people a carrier of the virus will infect in an entirely susceptible population. The phrase that is less talked about is Re, or the effective reproduction number – this is the average number of people a carrier of the virus will infect in the actual population.
R0 is estimated to be somewhere between 2 and 3. Re is calculated by multiplying R0 by the proportion of the population who have not yet been affected. The job of policy makers is not to reduce R0 to below 1, but to reduce Re to below 1 (this is the point at which the virus will eventually become eradicated). That can be done by washing hands, wearing masks and social distancing but…and here’s the good news…it will automatically happen over time. With each reported case, the proportion of the population who have not yet been affected diminishes – that means the number that you multiply by R0 diminishes, that means that Re diminishes.
Remember that to calculate Re we must first know what proportion of the population has not yet been affected and this is where the data is lacking. Currently coronavirus tests are only available in hospital, whilst at the same time official Government advice is that, ‘if you have Coronavirus symptoms you should not go to hospital’. Given it is estimated that anywhere between 20% and 80% of cases are asymptomatic and that you are told not to be tested even if you have symptoms, you can see why this data set is below par.
So official statistics (as at 26th March) reported that just 0.02% of the population had been affected. That leaves 99.98% unaffected and a worst case scenario Re of 2.99 – heavy handed policies are quite rightly needed to get this down to below 1.
However, a recent paper coming out of Oxford University saw experts suggest that more than 50% of the population had already been affected. That leaves 50% unaffected and a worst case scenario Re of 1.5 – that might imply little more than hand washing and good basic hygiene to avoid a medical disaster.
That leads me to the second problem in the data set and that’s the mortality rate…or more importantly the fatality rate. The mortality rate is easy to calculate and is the number of deaths divided by the total population. The fatality rate is calculated as the number of deaths divided by the number of cases; this brings us back to our earlier problem.
Looking at two countries, Italy and South Korea, highlights the problem. In Korea large parts of the population were tested ‘at random’ but in Italy only the worst symptomatic cases have been tested. In Italy just over 22% of cases were in people under the age of 49, in South Korea just over 60% of the cases were in people under the age of 49. In Italy the reported fatality rate is 10.2% in South Korea it was just 1.5%.
Policy appears to be being set to deal with a virus that will spread at a high rate with a high fatality rate but there is at least a chance that this is a virus whose spread has mostly already naturally been curtailed with a very low fatality rate. My conclusion is thus; I am glad that I do not have to make decisions whose impact is so large on so many based on such little information.
