Covid-19 blame game: Identify the usual suspects

By Surjit S Bhalla & Karan Bhasin

Related posts

EXCLUSIVE – Crypto giant Binance controlled ‘independent’ US affiliate’s bank accounts

June 5, 2023

Why investing just to save tax may be bad for you

June 5, 2023

India is going through a tragic Covid crisis, and our prayers are for those suffering in these frightening times. People are demanding answers. This is both fair and logical. We all would like to know what is going on, and the extent to which this crisis could have been prevented by timely proactive action. It is with this goal in mind that we revisited the issue we had last examined in some detail on January 16, 2021. Those were happy and happier times—with the Covid data available then, we concluded, somewhat prematurely optimistically, that India was approaching herd immunity. We were wrong.

Can we identify anything (in retrospect) that would have made the pain less, that could have prevented this explosive surge? India provides enough information on every “natural experiment” possible, e.g., different states going through different phases of the virus (waves and mutations) and we attempt to exploit that information. We do so via use of the Gompertz curve; also see bit.ly/2QjIP0B. The Gompertz curve is likely the most efficient, and most accurate, representation of a time-series process like the diffusion of a virus. It was developed in 1825 to study trends in mortality (and to make forecasts of the same).

As we all have witnessed, the world is literally littered with wrong assessments, and wrong forecasts, of the determinants of Covid19 and the optimal method to counter it. Infectious diseases are as old as humanity. What was new about Covid-19 was that countries, almost the entire world, chose the most extreme, and the most elitist method, of countering it—lockdowns. Despite its massive failure (see Lockdown Vs Covid; Covid Wins, bit.ly/3dzLJHs and COVID-19 India: Evolution and Performance bit.ly/3n5tay1), it is sad to see otherwise humane experts recommending lockdowns again—and doing so across the world. Think about it; more than 130 countries recommended and implemented lockdowns, but can anybody point to success? The usual “successful” suspects in this case are a handful of countries geographically close to the country of origin of the virus, China. The assumed success also includes countries as far away from China e.g. New Zealand. But less than 10 countries with “lockdown effectiveness” out of 200—and that is being recommended again? We agree with PM Modi who recently reiterated that “there is no substitute to testing, tracking and treatment.” This worked in India (and places like Viet Nam and Japan) and will work again, especially now that vaccines are available.

There might be more to the lockdown story. It also might just be a coincidence, but the people most in favour of lockdowns were (are) those in the political opposition. Media in the most media-rich, and media-savvy country—the US—went strangely silent about mismanagement of Covid within minutes after the close of polls on November 3, 2020.

The experience with Covid-19 should have taught us humility—the reality is that we just don’t know. We predicted herd immunity, and are surprised and shocked, with what is happening. Besides lockdowns, there have been several other favourite suspects. Let us take the example of seemingly the most intuitive recommendation to prevent infection—wearing of masks. The one organisation mandated to analyse Covid, the WHO, has made several missteps. It has been analysing flu epidemics for decades, and yet came out with a recommendation in November 2019—in the form of a detailed report—that masks were not very useful. Even as expert an expert as Dr. Anthony Fauci, the director of the National Institute of Allergy and Infectious Diseases, has had to reconsider, if not retract. In April of 2020, Dr. Fauci said that “there’s no reason to be walking around with a mask”; he changed his stance later as the pandemic progressed. Recently (late 2020), Dr Fauci advocated for the use of double masks and added that this was “common sense”.

So, what works? Possibly masks and social distancing—and meeting people in the open (we know, from painful experience, that lockdowns don’t work). In the recent India elections—noisy and in your face, and, therefore, ‘Covidly’ not correct—people have been wondering, and complaining aloud that the reason we have the new wave is because people have let down their guard, are no longer washing hands after touching metal (now, the WHO tells us we never had to do that!), and are not wearing masks, or keeping any social distance at political rallies, let alone the ‘scientifically’ proven six feet. By being a large, diverse country, India contains many natural experiments for statisticians to experiment with—and many conclusions to infer. Natural experiments include speculative frenzy that drove up the price of a NBFC that had Oxygen in its name even though the nature of business had nothing to do with supplying oxygen.

We report in the accompanying graphic a detailed analysis of all the large states in India, as well as an aggregation of small states. Actual cases (per ten thousand population) are reported in the second column. The Gompertz estimate (model estimated till January 31, 2021, to allow for out of sample forecasts assessment) is reported in the third column. The final column reports the percentage difference in the two. A priori, one would expect that the states which relaxed too early (letting activity happen) and/or states which held super-spreader events like election rallies, should have the largest deviation from what would be expected. In other words, states like West Bengal and Assam (especially the former) should have a massive uptick in “surprise” infections—the surprise being an excess over what was expected before the rallies, i.e., the percentage gap between actual and predicted cases.

The worst-performing state is Maharashtra—actual infections were 45% higher (as of April 17) than predicted. Punjab reports the second-highest deviation—42 %. This could be because of the farmer rallies without masks, but that took place in Delhi—a state which performs better (4% less infections) than expected. But Delhi has the highest incidence of cases—49 per thousand population. Maybe farmer rallies did cause a big uptick (but the glitterati was notably absent from objecting to the same, including international experts like Rihanna and Greta Thunberg). Kerala was supposed to be the best-performing state, and not too many BJP rallies there. It has the second-highest incidence (after Delhi), and its recent performance is only marginally better than average. Most of the “expert” complaints against election rallies have been aimed at where the BJP is holding the most rallies—West Bengal and Assam. Both these states show infections below that predicted before the rallies began in earnest. Their absolute infection rate is also low. What happened (or is happening)—maybe being outdoors (the opposite of lockdowns) is better than remaining indoors? We don’t know—but maybe we are beginning to understand.

A closer look at Bihar elections (conducted in October-November 2020) supports the above result. It shows the lowest number of cases (only 3) per thousand population. And the recent surge there is equal to that of Kerala. Choose your conclusion.

There is less than limited evidence to suggest that electoral rallies have resulted in an increased spread of the pandemic—and one may have to revisit this issue after a couple of months, once more data are available. Till then, armchair experts should introspect and appreciate India’s ability to conduct elections, with high voter participation, and do so in a pandemic.

Bhalla is executive director, IMF, representing India, Sri Lanka, Bangladesh and Bhutan. Bhasin is an independent economist
Views are personal and don’t necessarily represent the views of the IMF, its executive board, or IMF management