With most investors spending majority of their time reading and thinking about Coronavirus, it would be fair to say that their minds are now firmly “anchored” by the crisis. As Kahneman & Tversky showed 40 years ago, anchoring impacts our ability to think rationally. 300 years ago, Thomas Bayes created a pathbreaking theorem which can be used to rationally think about the risks around Covid-19. Applying Bayes’ Theorem shows that the risk of us dying from road accidents or from air pollution in an Indian city is at least FOUR TIMES HIGHER than the risk of us dying from Covid. Hence a rational investor like us cannot help being massively bullish on high quality Indian stocks.
“When the facts change, I change my opinion. What do you do, Sir?” – John Maynard Keynes
Wall to wall Covid coverage anchors us
Over the last month, the global media has gone into overdrive to give all of us nonstop coverage of the Covid–19 crisis. In addition, on social media, millions of self-proclaimed Corona gurus have emerged to spell out their visions of impending doom. In such circumstances, it is but natural that most people have got “anchored” to the belief that it is the end of the world as we know it.
The first psychologists to show us how anchoring works were the fathers of behavioural science, Daniel Kahneman and Amos Tversky. 40 years ago, they showed us how anchoring results in people taking a decision based on an initial view.
In one of their experiments they asked a group of high school students to give the product of the following equation within 5 seconds: 8 x 7 x 6 x 5 x 4 x 3 x 2 x 1. And to another group of high school students, they gave the following equation: 1 x 2 x 3 x 4 x 5 x 6 x 7 x 8. People tend to extrapolate the answers using only the first piece of information that they have and hence the median answer of the first group was 2,250 and for the second group it was 512. However, the correct answer to the equation is 40,320. (Link – pg, 1128, “Adjustment and Anchoring”)
Taking this simple example of anchoring and applying it to the current scenario, it appears to us that most people are overestimating the risk posed to them (and to their livelihoods and to their portfolios) by Covid-19. This overestimation is being fuelled further by tendency of people to resort to “confirmation bias” where people will give undue weightage to information that confirms their initial view.
Bayes’ Theorem to the rescue
Given the paucity of data available on Covid-19, how does one reach a realistic assessment of risk? We at Marcellus have taken a shot at it with the help of Bayes’ Theorem, which describes the probability of an event, based on prior knowledge of conditions that might be related to the event.
In simple terms, Bayes’ Theorem helps in calculating the probability of certain event (say, Event A) given that another event (say, Event B) has already happened. The formula is:
* Globally, the number of confirmed cases has reached 2 million and it is now showing signs of peaking (scroll down the page on this link and you will see the number of new Covid cases flat lining). However, it could well be that the number of new cases is levelling out because large parts of the world are under a lockdown and hence in our model, we have assumed 5x that number 10 million positive cases in India alone. We hope you will agree that this is an extremely conservative assumption.
# While it is still unclear about the distribution between symptomatic and asymptomatic cases, we have assumed that the number of asymptomatic cases is equal to symptomatic cases. We hope you will agree that this too is an extremely conservative assumption.
^ The total number of tests done so far in India is around 200,000 and the number of people tested positive for Covid in India is around 10,000. Hence the ratio of cases/tests is around 5% i.e. India seems to have one of the lowest Covid infection rates in the world (Link). Erring on the side of caution, we have assumed the infection rate to be 5% in our model. Hence the number of people having symptoms, given that the number of symptomatic cases is assumed to be 5m, is around 100m. [This number is likely to be an extremely conservative estimate because it feeds off the the two preceding assumptions which are extremely conservative.]
Note: Numbers are updated as of 14th April, 2020.
Even if one tests positive for Covid-19, the probability of a fatality is around 6% worldwide, while for India, fatality rate appears to be around 3% [Death rate = No. of deaths/no. of total cases, Link). This implies that under these extremely conservative (some will say cynical) assumptions made by us, India could see 300,000 Covid fatalities (10 million from the table above * 3%). To put this number of fatalities into perspective:
For those of us who live in Indian cities, you and I are at far greater risk of dying from a road accident or from air pollution in India than we are from Covid-19. In fact, the risk of us dying from road accidents or from air pollution in an Indian city is at LEAST FOUR TIMES HIGHER than the risk of us dying from Covid (and we are almost certainly overestimating the Covid related risks we face in India in the maths shown above). Of course, that shouldn’t stop us from scaring ourselves silly about Covid-19. However, the stock market rewards those who think about risks & rewards rationally, people like John Maynard Keynes, who other than being a pathbreaking economist was also a very good investor.
The data that is emerging from India is telling us that even amongst the 200,000 Indians who have been tested so far, the Covid-19 infection risks are amongst the lowest in the world. The data is also showing that the number of new Covid cases is already levelling out in India. We, as rational investors, believe therefore that to extrapolate the current lockdown in India and forecast a doom-laden future would be imprudent. In three separate notes recently, we have explained our view on how we are investing through the crisis:
Harsh Shah and Saurabh Mukherjea are Analyst and CIO respectively at Marcellus Investment Managers.
If you want to read our other published material, please visit https://marcellus.in/
Note: the above material is neither investment research, nor investment advice. Marcellus does not seek payment for or business from this email in any shape or form.