The Next Global Disaster Is on Its Way, and We Aren’t ReadyNiall Ferguson
The Covid-19 pandemic is not over, but it is already clear that Lord Rees, Britain’s astronomer royal, has won his 2017 bet with the Harvard psychologist Steven Pinker that “bioterror or bioerror will lead to one million casualties in a single event within a six-month period starting no later than Dec. 31, 2020.”
Last year, according to Johns Hopkins University, the SARS-CoV-2 virus claimed the lives of 1.8 million people. The global death toll could exceed 5 million by Aug. 1 — or 9 million, if one accepts the drastic new upward revision by the Institute for Health Metrics and Evaluation. It could have been worse, of course. In March 2020, some epidemiologists argued that, without drastic social distancing and economic lockdowns, the ultimate death toll could be between 30 and 40 million. Yet the cost of such nonpharmaceutical interventions has been enormous — for the U.S. alone, an estimated 90% of GDP.
Lord Rees’s was only one of many warnings before 2020 that humanity’s most clear and present danger was a new pathogen and the global pandemic it could cause. Yet somehow these warnings did not translate into swift, effective action in most countries when a pandemic struck. Why did so many democracies handle this crisis so badly?
The line of least intellectual resistance has been to blame populist leaders such as U.S. President Donald Trump, U.K. Prime Minister Boris Johnson, Brazilian President Jair Bolsonaro and now Prime Minister Narendra Modi of India. Certainly, they did not distinguish themselves, to put it mildly. In terms of excess mortality, however, Belgium fared worse last year than the U.S., the U.K. and Brazil. Yet its prime minister for most of 2020 was a liberal, Sophie Wilmes. Peru has been harder hit than almost any major country. Though its president, Martin Vizcarra, was also impeached (twice) last year, he cannot really be described as a populist.
Was democracy itself the problem? No. In China, the one-party state responded to the outbreak of the novel coronavirus in much the same way that its Soviet counterpart had responded to the 1986 Chernobyl nuclear disaster: with lies. On Dec. 31, for example, Beijing told the World Health Organization that there was “no clear evidence” of human-to-human transmission. The next day, eight Wuhan doctors who had sought to sound the alarm were detained. President Xi Jinping’s government prevented the spread of the virus beyond Hubei only by draconian restrictions on individual freedom.
Other authoritarian regimes fared worse, though we cannot be sure just how badly, as Russian and Iranian mortality statistics are not to be trusted. The true winners in their pandemic responses were Taiwan and South Korea, two East Asian democracies. The race to develop vaccines was won by biotech companies in the U.S. and Germany. The race to distribute them was won by Israel, the United Arab Emirates and the U.K.
We tend to draw a distinction between natural and man-made disasters. But a pandemic is made up of a new pathogen and the social networks that it attacks. We cannot understand the scale of the contagion by studying only the virus itself, because the virus will infect only as many people as social networks allow it to, and that in turn has a lot to do with politics.
Even an earthquake is only as catastrophic as the extent of urbanization along the fault line — or the shoreline, if it triggers a tsunami. A catastrophe lays bare the societies and states that it strikes. It is a moment of truth, of revelation, exposing some as fragile, others as resilient, and others as (to use Nassim Nicholas Taleb’s word) “antifragile” — able not just to withstand disaster but to be strengthened by it. In that sense, all disasters are man-made, in that our actions, including our political preparations and responses, determine the scale of the excess mortality.
“Strange Defeat” was the title the historian Marc Bloch gave his account of France’s collapse in the summer of 1940. In many ways, the American and European experiences of Covid-19 have both, in their different ways, been strange defeats, though it was germs not Germans that inflicted the casualties. Poor leadership played a part, no doubt. But it was certainly not Bloch’s contention that France’s strange defeat was all the fault of Prime Minister Paul Reynaud.
Disasters are by their very nature hard to anticipate. Some are “predictable surprises,” like Michele Wucker’s “gray rhinos” that we see rumbling toward us. Yet sometimes, at the moment they strike, these gray rhinos can metamorphose into Taleb’s “black swans” — seemingly bewildering events that it is now claimed “no one could have foreseen.”
This is partly because many disastrous events are governed by power laws, rather than a normal probability distribution of the sort that our brains more readily comprehend. Plotted on a graph, the distribution of pandemics is not the familiar bell curve, with most outbreaks clustered around the mean. Rather, if you plot the size of pandemics against the frequency of their occurrence using logarithmic scales, you get a straight line. The same is true of earthquakes.
This means that there is no average pandemic or earthquake; there are a few very large ones and a great many quite small ones, and there is no way of attaching a probability to the timing of a very large one. The same goes for man-made disasters such as wars and revolutions (which are more often disastrous than not) as well as financial crises — economic disasters that have lower death tolls but, often, comparably disruptive consequences.
A defining feature of history is that there are many more black swans — not to mention what Didier Sornette calls “dragon kings,” events so large in scale that they lie beyond even a power-law distribution — than a normally distributed world would lead us to expect. All such events lie in the realm of uncertainty, not of calculable risk.
Moreover, the world we have built has, over time, become an increasingly complex system prone to all kinds of random behavior, nonlinear relationships and “fat-tailed” distributions. A disaster such as a pandemic is not a single, discrete event. It invariably leads to other forms of disaster — economic, social and political. There can be, and often are, cascades or chain reactions of disaster. The more networked the world becomes, the more we see this.
Disaster management is made still more difficult by the fact that our political systems promote into leading roles people who seem especially oblivious to the challenges described above: subprime forecasters rather than superforecasters, to use the term coined by the political scientist Philip Tetlock.
The psychology of military incompetence was the subject of an excellent study by Norman Dixon; less has been written about the psychology of political incompetence as a general phenomenon. We can all readily think of individual incompetent politicians. But can we identify general forms of political malpractice in the field of disaster preparedness and mitigation? Five categories come to mind:
This is partly a problem of incentives. Leaders are rarely rewarded for what they did to avoid disasters — for the non-occurrence of a disaster is rarely a cause for celebration and gratitude — and more often are blamed for the pain of the prophylactic remedies they recommended.
Yet not all failures in disaster management are failures of leadership. Often the real point of failure is further down the organizational hierarchy. As the physicist Richard Feynman proved in the aftermath of the space shuttle Challenger’s destruction in 1986, the fatal lapse was not the White House’s impatience for a successful launch to coincide with a presidential address. Rather, it was the insistence of mid-level bureaucrats at the National Aeronautics and Space Administration that a risk of catastrophic failure was 1 in 100,000, while their own engineers put it at 1 in 100.
This, as much as blunders at the top, turns out to be a feature of many modern disasters. There is, as the Republican congressman Tom Davis said after Hurricane Katrina, a “vast divide between policy creation and policy implementation.” The way the Centers for Disease Control and Prevention botched testing in the early phase of the pandemic is a perfect illustration of the point. And does anyone even remember the name of the guy who was assistant secretary for preparedness and response at the Department of Health and Human Services last year? (Robert Kadlec is the answer.) As they say, he had one job …
In a disaster, the behavior of ordinary people — whether in decentralized networks or unruly crowds — can matter more than the decisions of leaders or orders issued by governments. What leads some people to adapt rationally to a new threat, others to act passively as bystanders, and others to go into denial or revolt? And why can a natural disaster end up triggering a political one, as disgruntled people form themselves into a revolutionary crowd? What causes a crowd to flip from wisdom to madness?
The answers lie in the changing structure of the public sphere. For a disaster is directly experienced by only a minority of people. Everyone else hears about it through some network of communication.
Even in the 17th century, the nascent popular press could sow confusion in people’s minds, as Daniel Defoe found when he researched the plague of 1665 in London. The advent of the internet has greatly magnified the potential for misinformation and disinformation to spread, to the extent that we may speak of twin plagues in 2020: One caused by a biological virus, the other by even more contagious viral misconceptions and falsehoods. This problem might have been less serious in 2020 had meaningful reforms of the laws and regulations governing the big technology companies been implemented beforehand. However, despite abundant evidence during the 2016 election that the status quo was untenable, almost nothing was done.
All disasters, in other words, are to some extent politically constructed, even if we think of some as natural and some as man-made. What should we do ahead of the next one? I have five suggestions.
First, we should stop trying to predict or even attach probabilities to disasters. From earthquakes to wars to financial crises, the major disruptions in history have been characterized by random or power-law distributions. They belong in the domain of uncertainty, not risk. It is better to admit that than to delude ourselves with unattainable and probably misleading precision.
Second, disaster takes too many forms for us to process with conventional approaches to risk mitigation. No sooner have we focused our minds on the threat of Salafi jihad than we find ourselves in a financial crisis originating in subprime mortgages. No sooner have we relearned that such economic shocks often lead to populist political backlashes than a novel coronavirus is wreaking havoc. What will be next? We cannot know. For every potential calamity, there is at least one plausible Cassandra. Not all prophecies can be heeded.
In recent years we may have allowed one risk — climate change — to draw our attention away from the others. In January 2020, even as a global pandemic was getting under way — as flights laden with infected people were leaving Wuhan for destinations all over the world — the discussions at the World Economic Forum were focused almost entirely on questions of environmental responsibility, social justice and governance, or ESG, with the emphasis on the “E.”
The dangers arising from climbing global temperatures are real and potentially catastrophic, but climate change cannot be the sole threat for which we prepare. Recognition of the multiplicity of threats we confront, and the extreme uncertainty of their incidence, would encourage a more flexible response to disaster. Not coincidentally, the places that did best in 2020 included three — Taiwan, South Korea and (despite a serious summer setback) Israel — that face multiple threats, including existential threats from neighbors.
Third, the more networked human society becomes, the greater the potential for contagion, and not just of the biological variety. A networked society needs to have well-designed circuit breakers that can swiftly reduce the connectivity of the network in a crisis, without atomizing and paralyzing society completely. Moreover, any disaster is either amplified or dampened by flows of information. Disinformation in 2020 — for example, viral fake news about bogus therapies or very safe vaccines — made Covid-19 worse in many places.
By contrast, effective management of information flows about infected people and their contacts helped contain the pandemic in a few well-run societies. The correct conclusion is not that big tech companies should have even more power to censor us and trace our movements. Rather, we need to learn from Audrey Tang, the minister who has pioneered the use of technology to empower Taiwanese citizens. That, not Xi’s Greater East Asian Surveillance State, should be the way of the future.
Fourth, Covid-19 exposed a serious failure of the public health bureaucracy in the U.S. and a number of other countries. The American epidemiologist Larry Brilliant, a key figure in the campaign to eradicate smallpox, has said for many years that the formula for dealing with an infectious disease is “early detection, early response.” In Washington and London, there was just the opposite.
Would a different kind of threat — say, a massive cyberattack on our critical infrastructure – produce an equally sluggish and ineffectual reaction? If the problems exposed by the pandemic are not specific to the public health bureaucracy but are general problems of the administrative state, then it probably would. How well would California cope with “the big one” on the San Andreas fault, to say nothing of the fires it would doubtless spark? I shudder to think.
Finally, there is a tendency throughout history, at times of acute social stress, for religious or quasi-religious ideological impulses to impede rational responses. We had all previously contemplated the danger of a pandemic, but more as entertainment (the movie “Contagion”) than as a potential reality. Even now, when other science-fiction scenarios are being realized — not only rising temperatures and climate instability but also the rise and expansion of the Chinese police state, to name just two — we struggle to react coherently and consequently.
In the summer of 2020, millions of Americans took to the streets of nearly 300 cities to protest loudly and sometimes violently against police brutality and systemic racism. However shocking the murder that precipitated the protests, this was risky behavior amid a pandemic of a highly contagious respiratory disease. At the same time, the rudimentary precaution of wearing a mask became a symbol of partisan affiliation. The fact that, in some parts of the country, gun buying seemed more popular than mask wearing testified to the potential for a public-order as well as a public-health disaster.
Covid-19 is not the last disaster we shall confront in our lifetimes. It is just the latest, after a wave of Islamist terrorism, a global financial crisis, a rash of state failures, surges of unregulated migration, and a so-called recession of democracy. Next up probably won’t be a disaster attributable to climate change, as we rarely get the disaster we expect, but some other threat most of us are currently ignoring.
Perhaps it will be a strain of antibiotic-resistant bubonic plague, or perhaps a massive Russian-Chinese cyberattack on the U.S. and its allies. Perhaps it will be a breakthrough in nanotechnology or in genetic engineering that has disastrous unintended consequences. Or perhaps artificial intelligence will fulfill Elon Musk’s forebodings, reducing an intellectually outclassed humanity to the status of “a biological boot loader for digital super intelligence.”