Tom Britton on the mathematics of the corona outbreak

21 Apr, 2020 at 19:53 | Posted in Politics & Society | 4 Comments

 

4 Comments

  1. Prof. Britton’s analysis throws no light on any major policy issues regarding COVID-19. This is because his analysis is grossly over-simplified with several fundamental deficiencies:
    .
    1. Neglect of Immunity
    Britton assumes that nobody who recovers from the virus develops any immunity. This is assumed in his formula for “epidemic size”, the proportion of population who are ever-infected (from 16’30” in the video).
    Relatedly, he fails to explain that “Herd Immunity” (no further growth in the epidemic) arises naturally when a fraction (1 – 1/Ro) of the population has developed immunity. This could be through natural infection and recovery from the disease, or (at a later date) through vaccination (if this ever becomes possible).
    .
    The same pessimistic assumption of zero immunity is made in his discussion of policies to reduce Re, the “effective reproduction number” (from 23’15” in the video). This unrealistic assumption leads to his advocacy of extreme measures to suppress the virus.
    He takes no account of the estimated 98% or more of infected people who recover from the virus, 90% with zero or only mild symptoms. Nearly all of these are likely to have immunity for several years at least.
    Britton assumes in the absence of major changes in people’s behavior, Re, the “effective reproduction number”, will remain constant at the initial Ro, the basic reproduction number = average number of new infections caused by a typical infection at an early stage in the epidemic. More realistically Re will decline quite quickly over time due to the development of natural immunity.
    .
    2. Britton makes no estimates of the catastrophic effects of the policies which he advocates on the physical and mental health of the population.
    In contrast to his assumption of zero immunity for individuals, he assumes that society and the economy are 100% immune to all of the harmful effects of the draconian measures imposed by governments as a result of the panic caused by forecasts from epidemiological models.
    .
    3. Failure to consider different policies for different risk groups
    Britton understates the serious weaknesses in his analysis due to the simplifying assumption of population homogeneity. He mistakenly suggests (from 20’50” in the video) that allowing for the differences between groups would only affect estimates of infections by 10-20%.
    (a) He fails to allow for the fact that the bulk of cases with severe adverse symptoms, hospitalisation and death stem from only about 10% of the population, namely the elderly and those with relevant pre-conditions. The other 90% have far better prospects when infected – most will have zero or only mild symptoms.
    (b) He advocates lockdowns, quarantines etc. for the whole population until the virus is suppressed. A more sensible policy would distinguish between high and low risk groups:
    – Vigorously provide generous and humane protection/quarantine arrangements on a voluntary basis for high risk groups. Those who choose to take the risk of infection and death would be free to do so.
    – Allow the other far greater majority 90% of the population the freedom to resume normal life a.s.p. and develop natural immunity. The timid would still be able to self-quarantine if they so wish.

    • It might be good for economics to discriminate against “the elderly and those with relevant pre-conditions” and “Allow the other far greater majority 90% of the population the freedom to resume normal life a.s.p. and develop natural immunity.”
      But that depends on how the virus develops. And what’s good for economics is not always good for people.
      This we can se from the lack of preparedness for stopping epidemics which has been created by excellent neoliberal and monetary economics. Which has already made the pandemic much more costly than it would have been if the necessary preparations had been made. In both lives health and money.

  2. I agree with much of the critique put forward here — and it actually reminds me not so little of the critique I myself many times raise against over-simplified economic models. On the other hand, I think Tom is very transparent about his assumptions and lack of substantive knowledge of the particulars of this specific epidemic.

  3. It is ironic that Kingsley Lewis defends mathematical economic models that include homogeneity (or one representative agent), but attacks this epidemiological model on the homogeneity assumption, because he does not like the conclusions reached in this case. And Professor Syll seems happy to accept the assumptions he attacks in economic models, or at least thinks the conclusions reached in the epidemiological model here are valid. The epidemiological model presented is ergodic …
    .
    It seems one starts with a conclusion, then picks and chooses models and criticisms that validate one’s pre-formed conclusions. Math models are good here despite unrealistic assumptions because I like the conclusions, but the unrealistic assumptions make math models bad there because I don’t like the conclusions reached.
    .
    The old saying “a pox (!) on both their houses” comes to mind; we should listen neither to epidemiologists nor economists when making public policy.
    .
    That being said, I would agree with Kingsley’s pandemic policies as long as everyone was paid a decent basic income, so they could choose to lock themselves down (or escape to the woods) without disastrous economic consequences.


Sorry, the comment form is closed at this time.

Blog at WordPress.com.
Entries and comments feeds.