Abstract: We depend on our national leaders to guide us through pandemics such as COVID-19. In the lead-up to the pandemic a number of world leaders were casual in their response, scoffed at the outcomes and did not take the preventive responsibility seriously. This article explores on why our leaders, and experts they depend on have tended to be wrong, what were their underlying biases that caused these decisions to a disastrous effect. It then sets our we can use modern judgement frameworks by to provide better judgement under situations of extreme uncertainty such as the coronavirus pandemic. The figures quoted in this article were accurate at time this was published i.e. 6th April 2020
A disastrous prediction
Let’s go back in history about 30-years ago. The top consulting company in the world, McKinsey was asked by AT&T, the largest telecom provider in the United States to predict the global market for mobile phones. McKinsey came back to AT&T and told them that the total addressable market globally was 900,000. They couldn’t have been more wrong. McKinsey may have gotten away with this and gone on to pursue other lucrative projects; it did, however, leave AT&T out of a vast market opportunity to disastrous effect.
The same consulting firm has just released a report on COVID-19 with a simulation of scenarios. Now, why would you believe them? McKinsey could not foresee potential changes in technology and adoption trends and was not the expert in predicting the market potential of mobile phone adoption. Their misprediction could have been as a result of a number of causes such as narrow framing and confirmation, overconfidence, confirmation bias – all covered in the noble laureate Daniel Kahneman’s masterpiece Thinking Fast and Slow. It is for the same reasons you should not believe them or any single expert in making long-range predictions of a highly uncertain future of pandemic spread. PS: If you have a look at the number of scenarios predicted, United States does not even figure in them, despite the fact it has the largest number of cases today
Experts are wrong 40% of the time, and overconfidence plays a significant role
In a study, doctors who were sure about their diagnosis turned out to be wrong 40% of the time. The consequences of this are enormous when it comes to life-threatening conditions. When students predicted that they had only a 1% chance of being wrong, they were wrong 27% of the time.
Experts and non-experts alike are overconfident about their predictions and often wrong.
For uncertain and complex events, there is no reliable expertise
The future has an uncanny ability to surprise, and COVID-19 is no exception. It’s a black swan event, a butterfly effect. An outbreak in China has brought the world economy to a standstill. At the time of writing, we have a semblance of coordinated action. However, a few weeks ago, we had presidents and prime ministers making casual statements about the coronavirus – President Trump even predicting the US would be back to normal by Easter (i.e. in less than 2 weeks time).
Risk and uncertainty are two different things. Risk is heading into a known unknown; we can estimate the likelihood of occurrence of an unpredictable (“risky”) event, usually helped by historical data and frequency of occurrence. We can also plan for taking measures to mitigate against it. Uncertainty can’t be predicted, and the complex systems, the spread of contagion, and lack of precedent make predictions unreliable.
Even if experts get it right, our leaders who often rely on experts get it wrong.
Our leaders were advised by experts, who are unfortunately not…
While we are waiting to see what is unfolding on coronavirus, here’s what experts have said. Trump not so long ago said that this was no more than a flu and more people die of flu and this is nothing to worry about it.
If you don’t believe the world leaders changing their stance, have a look at this tweet and video from Boris Johnson where he was flippant about coronavirus.
From @PoliticsJOE_UK Britain was given a head start to tackle the coronavirus as it spread through other countries. Much of that time has been wasted.
On 28th February 2020, Trump downplayed the virus, saying it will disappear one day like a miracle. The number of cases then was 63. Trump, similarly, disregarded the WHO figures on 5th March 2020 saying his “hunch” was that death rate was significantly lower.
See Trump’s quote to fox news in the footnote. At the time of posting this (on 3rd April 2020) the number of cases in the USA is over 265,000 and over a 1 million cases worldwide.
Why were the leaders so optimistic?
One of the problems is optimism and overconfidence and reliance on one’s own opinion. A second problem is the human inability to understand exponential growth. The number of cases in the US was 63 on 28th February, less than 30 days later its crossed a 100,000 mark. Paul Graham, the founder of YCombinator, one of the world’s most successful accelerators, had a wonderful quote on twitter,
‘People aren’t surprised when I tell them there are 13,000 Covid-19 cases outside China, or when I tell them this number doubles every 3 days. But when I tell them that if growth continues at this rate, we’ll have 1.7 million cases in 3 weeks, they’re astonished.’
Here’s a simple thought experiment:
Let’s take lake Como in Italy that has an area of 146 sq km, which is about the size of Preston in the UK or the city of Jaipur in India. Now, imagine a swarm of lily pads that doubles every day. At some point, the lilies would cover the entire area of the lake. Let us say it takes 30 days to cover the whole lake. If you flew over lake Como on day 29, you would see half the lake hidden with lily pads and half covered by the body of water. On day 28th it would ¼ covered with lily pads and so forth. Still substantial. Now imagine that you flew over this very lake on day 10, the lilies were doubling at the same rate to cover the entire lake in 30 days.
Here’s what you would notice on day 10. Nothing! The lilies would only cover 0.000139 sq km or 139 sq meters, the surface area of an apartment in Jaipur, unnoticeable from a plane. The trouble with exponential growth is not when you see it coming accelerating (in which case measures are too late), but how slow it is in the beginning, that it’s almost unnoticeable. And that’s what makes it hard for leaders to imagine future consequences. At the end of the article, you can see a link to a Washington Post that shows the visualisation of the “rapid” spread of contagion based on different strategies that a country deploys.
Biases have a strong role in misleading us
In highly changeable, high-stress environments, such as Covid-19, decision-making biases are likely to appear. In this period, where decision making is essential, biases can have a damaging effect. Tom Davenport, in his MIT Sloan review article, talks about the outcomes of emotion-driven beliefs and intuition that are powerful at guiding people toward less-than-optimal decisions. Some of these biases that effected the Covid-19 decision-making process are Status-quo, political bias, confirmation bias and the availability heuristic, well worth a full read of Tom’s article here.
This is not a black swan event
Nassim Taleb, the author of the 2007 bestseller “black swan”, co-authored a paper in late January when the virus was mainly confined to china and cautioned that that, owing to “increased connectivity,” the spread will be “nonlinear”. A black swan event is rare, unpredictable, catastrophic events, this pandemic was predictable and a matter of probabilities. He went on suggest that for statisticians, “nonlinearity” describes events very much like a pandemic: an output disproportionate to known inputs (the structure and growth of pathogens, say), owing to both unknown and unknowable inputs (their incubation periods in humans, or random mutations), or eccentric interaction among various inputs (wet markets and aeroplane travel), or exponential growth (from networked human contact), or all three. In other words, these are fat-tailed processes making conventional risk management techniques inadequate.
So, what made the world leaders change their opinions?
Once the leaders take a position, they tend to seek information that confirms their conviction at the cost of ignoring information that provides contrary evidence. They select ‘experts’ accordingly. A necessary leadership trait is a need to be seen to be consistent. Usually, it is a jolt, when a leader needs to change his or her opinion, change the narrative and develop a new strategy.
Different countries reacted differently. In the USA, it was not the briefings by the Centre for Disease Control (CDC), but an experience Trump had, up close and personal, on a visit to Florida to meet senators where a number later turned positive. On 14th March, Trump was hosting the Brazilian president at a weekend getaway at Mar-a-Lago. The US President was in contact with several people who were carrying the virus.
As they say, when your friend has lost his job, it’s a recession when you lose your job, it’s a depression. This personal experience probably got Trump into gear and within 48 hours or so he had mobilised a whole range of measures, including building a public-private partnership. This is a typical example of someone moving from one position to a completely different one.
The UK government shifted its stance on 17th March 2020 in light of scientific advice that’s been shaping its response to the coronavirus pandemic, which included 30 documents, including academic papers and the influential Imperial College London modelling. This update led to a dramatic change in strategy, from mitigating the disease to trying to suppress its spread. India, who had been avoiding testing, to prevent panic according to the Director of the Indian Council of Medical Research, came under pressure from WHO that made the Indian Prime Minister take one the most aggressive draconian measures to lock down the economy entirely.
Sometimes it is a credible internal source as in case of the UK; sometimes it’s a personal experience that trumps statistics (pun intended) as for the US and sometimes its external pressure, as in the case of India, which had avoided testing.
You can see the simulation of the Washington Post and how this particular disease spread depending on the strategies people deployed on how close people came into contact, and you can see that is very slow at the beginning and then suddenly it takes off.
So how should we act under uncertainty in the spirit of what we hold and believe to be true?
Well, expertise can be divided across the range, and it’s hard to believe what is right or wrong, partly also because in uncertain events there is no precedent. Any simulations or scenarios are based on conjectures of things that might follow.
We need to make judgements and trust in a process that we can follow reliably.
In cases of uncertainty like this, we need to focus on the process of making our own decisions based on our own beliefs, preferences attitude to risks and short-term goals. It is hard to plan far in advance, even governments are taking short term measures and evaluating at the end of these cycles.
The HBR article Elements of Good Judgement, by Sir Andrew Likierman, supports this. He talks about the process that takes into a number of factors such about ‘what is my source of information’, and ‘what and who did I trust’, ‘what else do I know about this’, ‘what do I feel I believe’ and then creating choices. He explains that you can’t make long-term decisions under these circumstances.
A colleague came to me three weeks ago asking for advice, when the pandemic was building up and a lockdown was imminent. Should I stay in the UK or should I go to India? He had done a classical spreadsheet analysis on the pros and cons, numbers today, rate of growth, hygiene governments ability to respond etc. His final decision was based on what he believed and what his gut told him. He did not believe the numbers India was declaring were real (yes India was not conducting sufficient testing as mentioned earlier), and he did not have the confidence that if the situation got worse the government and structures in India would hold out. He took a gut decision and stayed in the UK.
Clear short-term decisions, that does not rule you out of options in the future, and decisions that create future options are likely to lead to better outcomes both in short and medium terms that long-range decisions.
Photo credit: NeONBRAND on Unsplash
#coronavirusimpact #covid19, #executiveeducation #changeschool, #pandemicpreparedness, #fightagainstcoronavirus, #StayHomeStaySafe
 From the Economist.com “HERE is a cautionary tale about a telephone giant and a management consultancy…” https://www.economist.com/special-report/1999/10/07/cutting-the-cord
 Thinking Fast and Slow, see Daniel Kahneman’s analysis of narrow framing (p. 87), overconfidence (pp. 199–201, 209–12, and 259–63), confirmation bias (pp. 80–84), and emotion and indecision (pp. 401–6).
 The 40% failure rate is described in Brooke Masters, “Rise of a Headhunter,” Financial Times, March 30, 2009, http://www.ft.com/cms/s/0/19975256-1af2-11de-8aa3-0000779fd2ac.html#axzz2401DwtbW.
Heath, Chip. Decisive: How to make better choices in life and work (p. 254). Random House. Kindle Edition.
 Robin Hogarth (2001), Educating Intuition (Chicago: University of Chicago Press), pp. 218–19.
 Decisive by Chip & Dan Heath Chapter 1, The four villains of decision making
 There is usually a host of factors that are currently unknown but that are in fact knowable—that could be known if the right analysis were done. HBR Article December 1997 on Strategy Under Uncertainty by Hugh Courtney , Jane Kirkland and Patrick Viguerie.
 On MSNBC news, aired on 28th February 2020. https://www.msnbc.com/deadline-white-house/watch/trump-coronavirus-will-disappear-one-day-like-a-miracle-79636549723
 Trump: Trump, however, told Fox viewers that the death rate was even lower — a “fraction of 1 percent” — based on his “hunch.” Here’s the rest of the quote:
Now, this is just my hunch, but based on a lot of conversations with a lot of people that do this, because a lot of people will have this and it is very mild… So if, you know, we have thousands or hundreds of thousands of people that get better, just by, you know, sitting around and even going to work, some of them go to work, but they get better and then, when you do have a death like you had in the state of Washington, like you had one in California, I believe you had one in New York, you know, all of a sudden it seems like 3 or 4 percent, which is a very high number, as opposed to a fraction of 1 percent.
 This change is not just pivot, in the parlance of the start-up way.
 New York times, 14th March https://www.nytimes.com/2020/03/14/us/politics/trump-coronavirus-mar-a-lago.html
 From MIT technology review https://www.technologyreview.com/s/615377/coronavirus-model-london-vaccine-suppression-imperial-deaths/)
 BBC News: India grossly undertesting https://www.businesstoday.in/latest/trends/india-grossly-under-tested-on-coronavirus-urgent-steps-needed-to-ramp-up-testing/story/398488.html
 HBR Elements of good judgement- Sir Andrew Likierman, January 2020 R2001H-PDF-ENG
Viren Lall is the Managing Director at ChangeSchool, a specialist in entrepreneurship and a an adjunct (visiting faculty) professor of management. He has delivered programmes in 15 countries and manages international partnerships across the 6 GCC countries, South Asia and Central Asia. His contributions on organisational change were published by Kogan Page. He has launched two start-up companies in the past. He has advanced degrees in computer science and engineering from King’s College London, IIT Delhi and has an MBA from London Business School.