The coronavirus pandemic has highlighted the difficulties of managing risk in an increasingly globalised and interconnected world. We explore how organisations can stay resilient through an era of radical change.
Pavia, located 20 miles south of Milan, looks much like any other small Italian town. Cobbled lanes wind between fashionable boutiques; medieval ruins sit alongside high-rise office blocks. But this unassuming place has an impressive claim to fame.
Pavia is the birthplace of Gerolamo Cardano, one of the brightest minds of the Italian Renaissance. A physician, astrologer and inveterate gambler, Cardano is best known for his treatise Liber de Ludo Aleae (Book on Games of Chance), which used mathematical formulae to predict the results of successive dice throws. Written in 1525, it was the first systematic attempt to define the principles of probability.
Though Cardano didn’t realise it – he was simply trying to improve his luck at the gambling table – his book laid the groundwork for the study of risk. As probability theory developed, it enabled people to define future possibilities and choose alternative courses of action, taking into account the likelihood of success or failure. This approach revolutionised whole spheres of human activity, from warfare to wealth allocation, from farming to family planning.
Cardano’s legacy is evident in Pavia’s unusual concentration of betting shops and lotto machines. In recent months, his name has cropped up repeatedly in conversations among the town’s citizens and officials, as they grapple with the implications of the coronavirus outbreak.1 Is it safe to visit family during lockdown? When should schools and workplaces reopen? The answers to such questions depend on weighing the odds.
Pavia’s residents are not alone. Around the world, the COVID-19 pandemic has made dodging risk the stuff of everyday life. People have been forced to make tough decisions about how best to protect their health and safeguard their livelihoods. As a recent Financial Times op-ed put it: “We are all risk managers now.”2
But the pandemic also shows some events are simply impossible to predict with any certainty, however sophisticated our understanding of risk. In a highly interconnected world, a viral outbreak at a Chinese wet market can rapidly escalate into a global crisis. Predicting the shape of the future is more difficult than ever.
Some experts believe heightened uncertainty may be the new normal. In their book Radical Uncertainty: Decision-making for an unknowable future, economists John Kay and Mervyn King argue the modern era is characterised by “uncertain futures and unpredictable events” that confound probabilistic risk models.3
We simply don’t know how this virus will evolve and what the economic consequences are going to be
COVID-19 is “absolutely an example of radical uncertainty”, Kay tells AIQ. “We simply don’t know how this virus will evolve and what the economic consequences are going to be. Our argument is that we should stop pretending to have more knowledge about the world than we actually do.”
This doesn’t mean we must simply accept our fate, however. As we’ll discover in this article, those who can stay informed, flexible and resilient will find opportunities, even during times of extreme change. In part one, we examine how theories of risk have evolved, from the gambling dens of medieval Italy to the modern trading desk. In part two, we identify some guiding principles for investors hoping to plot a course through an increasingly unpredictable world.
Risk and uncertainty
For most of history, the future was considered to be beyond human knowledge or control. As Peter Bernstein wrote in his classic study Against the Gods, “the future was a mirror of the past or the murky domain of oracles and soothsayers”.4
We realised our actions can influence the future; it is therefore not predetermined
What changed was the advent of risk. We realised our actions can influence the future; it is therefore not predetermined. The effects of human behaviour can be understood and modelled by drawing inferences from observable data.
Cardano’s statistical method, developed by later mathematicians such as Blaise Pascal, Pierre de Fermat and Carl Friedrich Gauss, formed the basis for probabilistic risk calculation. As our understanding of risk became increasingly sophisticated, it enabled massive leaps forward in science, engineering, government and finance, as individuals and organisations began to plan and strategize more effectively.
The early thinkers on risk knew the limitations of their theories. Not every situation can be quantified using statistical methods, which are best suited to scenarios with fixed rules and parameters, like games of cards or dice.
In the 20th century, economists Frank Knight and John Maynard Keynes drew attention to the limits of probabilistic analysis in their writings on finance. They distinguished between risk, which can be measured, and uncertainty, which cannot.
As Keynes wrote in an essay summarising his landmark text, The General Theory of Employment, Interest and Money (1936): “I do not mean to distinguish what is known for certain from what is merely probable. The game of roulette is not subject, in this sense, to uncertainty. The sense in which I am using the term is that in which a European war is uncertain, or the price of copper and the rate of interest 20 years hence…about these matters there is no scientific basis on which to form any calculable probability whatsoever. We simply do not know.”
Pangolins and black swans
To better understand the distinction between risk and uncertainty, consider some other examples. Credit investors know there is a risk a company may default on its debt: this can be estimated and priced by looking at the issuer’s credit rating, strength of its balance sheet, quality of its governance and the state of the wider economy. This risk is a “known known”.
But there are other hazards that can’t be measured precisely, or even foreseen. This is the domain of uncertainty: the “unknown unknowns” that complicate more linear risk assessments.
Black swans are outliers, because nothing in the past can convincingly point to their possibility
One of the earliest uses of this phrase comes from the aerospace industry. In 1954, two de Havilland Comet passenger jets crashed in mysterious circumstances, stupefying engineers and safety officials. After an exhaustive investigation, the cause of the accidents was found to be metal fatigue originating in the corners of the jets’ square-shaped windows. (This is the reason most plane windows are now oval in shape.) Engineers described the square-window problem as an unknown unknown, or “unk unk” for short – a design flaw so tiny it would have been impossible to spot before it cascaded into a crisis.5
Derivatives trader-turned author Nassim Nicholas Taleb used his famous metaphor of “the black swan” to describe unknown-unknown events. In his definition, black swans are outliers, because nothing in the past can convincingly point to their possibility – even if it is tempting to concoct explanations for them afterwards.
True black swans are rare, however. More often, risks belong to a related category, “known unknowns”: hazards we can anticipate in a general sense without being sure exactly when or where they will occur.
COVID-19 is a good example. Experts warned repeatedly of the threat of a pandemic in recent years. In a TED Talk in 2015, Microsoft founder Bill Gates said global health systems “are not ready” for the outbreak of a novel flu pathogen.6 As recently as June 2019, Politico magazine modelled the course of a future coronavirus pandemic and analysed its likely effects on public policy.7 But no-one predicted COVID-19 would emerge in late 2019 in the Chinese city of Wuhan, jumping the species barrier from a wild animal – probably a pangolin or a bat – to a human host. Nor did anyone foresee that the spread of the new virus would coincide with the sudden decision by the Saudi government to slash the price of oil, exacerbating the market panic.
“Few people, at least outside of the epidemiological experts, would have countenanced a world where social distancing and the cessation of freedom of movement would become reality in such a short space of time,” says Mark Robertson, head of multi-strategy funds at Aviva Investors. “In addition, the decision by Saudi oil producers to enter into a price war was the last thing an already fragile market needed.”
Here be dragons
The coronavirus outbreak has prompted intense debate about the uses of big data and artificial intelligence in monitoring systemic threats. Could a sophisticated algorithm have anticipated the emergence of a deadly virus, or at least modelled its impact once it emerged?
Didier Sornette, a professor at the Swiss Federal Institute of Technology Zurich, is one of the world’s leading experts on risk. By carefully monitoring disturbances in complex systems, he has been able to detect early signs of crisis in various environments, from heavy industry to finance (see The Scientist: Didier Sornette).
Sornette’s theory is that extreme events – or “dragon-kings” – often announce themselves through minor shifts that can be discerned by experts who know what to look for. For instance, Sornette was able to predict the failure of a pressure tank in an Ariane space rocket by monitoring the progress of miniature acoustic emissions deep in the tank’s matrix of carbon fibres. He has used similar methods to predict seismic aftershocks, landslides and the formation of some financial asset bubbles.
Not even the most advanced data-driven methods could have anticipated the onset of COVID-19
But Sornette appreciates many events are impossible to capture within numerical models. Not even the most advanced data-driven methods could have anticipated the onset of COVID-19, though they might have been helpful in mapping its subsequent transmission across borders. “Predicting the original case – ‘patient zero’ – would have been impossible,” he says.
Sornette’s approach acknowledges the difference between risks that can be modelled probabilistically and uncertainties that cannot. Though the distinction is simple enough, it is often overlooked by experts who believe they can perfectly chart the course of the future.
As Sornette puts it: “I like to say that nature is more imaginative than mathematicians, physicists, engineers, specialists of all kinds. We are very often taken by surprise when a catastrophe occurs, as the path to it has not usually been imagined.”
The trouble with models
Such humility is not always evident in "expert" forecasters, who are taken by surprise more often than they may like to think. A famous study by the US psychologist Philip Tetlock found the average expert in geopolitics and economics –defined as those with more than 12 years’ experience – is about as accurate in predicting the future as a chimpanzee throwing darts at a target.8
Tetlock discovered amateurs often perform better than professional futurologists. His government-funded Good Judgement Project unearthed a clutch of so-called “superforecasters”, whose predictions consistently beat the average, at least when it came to answering short-term questions – up to about 400 days into the future – with highly constrained parameters.
Superforecasters come from a range of backgrounds, but share a particular character trait: open-mindedness
Superforecasters come from a range of backgrounds, but share a particular character trait: open-mindedness. They are rarely wedded to a single ideology or perspective; they are open to challenge and debate in the interests of learning more; and when the evidence shifts, they are willing to change their minds.
By contrast, experts in technical fields often cling to a damaging sense of certainty about the future. This can leave their organisations vulnerable to events that lie outside the scope of their models.
“People in government, people in decision-making positions in corporations, want levels of certainty that models purport to provide,” says Margaret Heffernan, author of the new book Uncharted: How to map the future together. “The problem is that all of the real risk – the systemic risk – appears to go away, and the possibility of picture-perfect decisions starts to feel available.
“The truth is since every single forecast can only have probabilities attached to it – and those probabilities will always be under 100 per cent – the opportunity to make the perfect decision is always elusive. We have to make trade-offs and try to make the best decisions we can with the information we have, but that information will keep changing, and very few models keep up with that pace of change.”
Kay and King make a similar point in Radical Uncertainty: “Instead of recognising uncertainty and adopting policies and strategies that will be robust to many alternative futures, banks and businesses are run with reliance on models which claim knowledge of the future that we do not have and could never have.”
Finance is a prime example. Over recent decades, hedge funds and investment banks have built complicated models that purport to track all manner of commercial and financial risks, including those that are ill-suited to probabilistic analysis. The distinction between risk and uncertainty has been elided.
The industry’s reliance on these kinds of models was dramatically illustrated during the global financial crisis. At the height of the turmoil, Goldman Sachs’ then-chief financial officer David Viniar famously said the company’s risk model indicated markets were undergoing “25-standard deviation moves several days in a row”.9
The probability of a single 25-standard deviation event is so low it would take up too much space to represent numerically on this page. Suffice to say, it is equivalent to the chances of a single person winning the UK National Lottery 22 times consecutively.10 Given those odds, the universe has probably not existed long enough for there to have been several days on which 25-standard deviation events could occur. Far more likely is that the model was flawed.
One common reason for the failure of probabilistic risk models is that the past is often an imperfect guide to the future. Banks that fed their models with data drawn from the relatively calm decades of growth and prosperity that followed the end of the Second World War were in for a nasty surprise when the financial crisis hit.
The lesson is that you cannot derive a probability about the world from a probability that’s developed in a model
“The lesson is that you cannot derive a probability about the world from a probability that’s developed in a model,” says Kay. “The database with which Goldman Sachs built its model came from a period in which banks didn’t go bust.”
Another issue is that risk calculations have become more complicated in the context of a globalised economy, in which a single event can trigger domino effects across multiple countries and markets. New factors enter the equation, feedback loops accumulate, and linear risk events quickly spiral into the domain of uncertainty.
The butterfly defect
According to Taleb, financial crises are becoming more damaging because of the physical and technological connections that characterise the modern world. Such connectivity increases the occurrence of “fat tails”, named after probability distributions that show unexpected thickness at the extreme end of the bell curve.11
Put simply, there are now more situations in which a single variable – a virus, asset bubble, cybersecurity failure, natural disaster, geopolitical spat – can have outsized effects. Under such conditions, quantifiable risks are often shadowed by unknowable uncertainties.
Consider the supply chain for a product such as Apple’s iPhone, which links high-end Korean chipmakers, Chinese manufacturing facilities and thousands of small, specialist companies that contribute different components to the finished machine. A single interruption at any point in this highly efficient and finely tuned process can result in delays, supply constraints and price increases further down the line.
A known-unknown event – the coronavirus pandemic – has amplified the risks inherent in this intricate system. Central China, where COVID-19 originated, hosts a cluster of manufacturing firms, including Hon Hai, Apple’s main supplier; similarly, Gumi Industrial Complex just outside Daegu, the city at the centre of South Korea’s coronavirus outbreak, produces most of the world’s memory chips and LED displays. The virus-related cessation in work at these facilities is expected to lead to at least a ten per cent fall in global smartphone shipments this year, and knock-on impacts are already being observed across a range of companies and industries.12
Supply chains are so integrated and efficient these days, there is less flex when there is an issue in one part of the world
“Supply chains are so integrated and efficient these days, there is less flex when there is an issue in one part of the world,” says Alistair Way, head of emerging market equities at Aviva Investors. “There is no easy way Apple can shift iPhone production away from Hon Hai because it is so efficiently set up with customised facilities.”
We are all familiar with the butterfly effect, the metaphor derived from chaos theory that suggests an insect could flap its wings and cause a tornado a thousand miles away. In a 2014 book, University of Oxford economist Ian Goldin coined a new phrase, “The Butterfly Defect”, to emphasise the inherent fragility of a deeply interconnected world.13
“Everything we have seen since  shows the concerns I expressed then about systemic risk in systems – be they health systems, financial systems, infrastructure systems – have proved themselves to be true. COVID-19 is just the latest example,” Goldin says. “Sadly, we are not seeing a full appreciation of the full implications of this. I don’t see anything in the direction of travel in health or in infrastructure that is going to make us more resilient.”
As new technologies such as Internet of Things introduce still more connections to the global economy, risk management will have to evolve quickly, argues Warren Black, an expert on the dynamics of complex systems and founder of Complexus, a risk consultancy.
“All our risk management standards assume risks happen in a logical, sequential, cause-and-effect way. As our systems and environments get more and more complex, that’s no longer true; at the highest level of complexity, you have chaos. Nothing can be predicted or proactively controlled when there is chaos, so conventional risk management techniques don’t work in environments of advanced complexity,” says Black.
The challenges of managing risk in this new era of radical uncertainty may be daunting, but they are not impossible to overcome. Organisations that can grasp the unpredictable nature of the modern world – and recognise the limits of their own knowledge – could find opportunities to thrive.
Attempts to predict the future with certainty are doomed to failure
One key lesson to draw from the events of recent months is that attempts to predict the future with certainty are doomed to failure. This is especially the case in finance, where timing the market depends on being right in the specifics, not the generalities.
Imagine that a portfolio manager had listened to Bill Gates’s warning of a pandemic in 2015 and repositioned their fund defensively in expectation of imminent market disruption. They would have missed out on five years of outsized returns as equities soared, even if recent events proved Gates was broadly correct. Clients would undoubtedly have questioned such a conservative strategy – particularly given the performance some passive funds were able to deliver over that period thanks to their unmoderated stock-market exposure.
A more sensible response to uncertainty would be to build portfolios that stand to perform well in a range of scenarios, argues Euan Munro, CEO of Aviva Investors.
“It would be a tall ask to say an asset manager should have had a research team that would have known a virus was going to come and spook the markets to this extent. Far less forgivable would be not having a portfolio that had some generic resilience to extremely disruptive events – whether that’s a collapse of a banking system, a health crisis or a geopolitical issue,” he says.
Plan for alternative futures
In Radical Uncertainty, Kay and King argue the best way for investors to stay resilient and flexible in a hyperconnected world is to plan for “alternative futures” through the adoption of “multiple strategies”.
Their exemplar is an unconventional oil-company executive named Pierre Wack, a former journalist and student of Indian mysticism who became famous in risk-management circles for his work at Shell in the 1960s.
Wack threw out the centralised planning model the company had previously used and encouraged his teams to think outside the box in planning for a range of potential futures. Long before the creation of OPEC, he speculated about the risk major Middle Eastern energy producers would form a cartel to exert monopoly power. As a result, Shell was able to weather the oil crises of the 1970s much better than its competitors.
Scenario planning has since been applied in a range of other contexts, from business to policymaking. And the approach has a close analogy in modern asset management, where resilience depends on building diversified portfolios that can withstand a range of possible developments.
You need to have a way to stress test alternate scenarios, explicitly for "shocking" your portfolio through those environments
“In all portfolio construction processes, you need to have a way to stress test alternate scenarios, explicitly for ‘shocking’ your portfolio through those environments,” says Josh Lohmeier, head of North American investment grade credit at Aviva Investors. “That helps you reallocate or rescale those ideas in a way that allows you to capture the inherent alpha while simultaneously acknowledging there will be periods of volatility. There will be exogenous shocks to the market that cannot be predicted; you need to prepare for those every day.”
This method of portfolio construction isn’t based on predicting low-frequency events, but about building in protection against categories of foreseeable risk.
Take the supply-chain example: investors need not have anticipated the outbreak of a new strain of flu to have discerned the vulnerability of the complex smartphone supply chain to a sudden catastrophe. A natural disaster, terrorist attack or geopolitical incident – such as a worsening of the US-China trade war – might have triggered similar disruption.
Preparing for multiple scenarios inevitably means allocating time and resources to model the impact of events that never come to pass. But such portfolios should benefit from the quality known in engineering as “modularity”, in that a single failure should only affect discrete parts, not the system as a whole.
“The coronavirus sell-off demonstrates the truth in a statement sometimes attributed to Mark Twain: ‘It ain’t what you don’t know that gets you into trouble, it’s what you know for sure but just ain’t so,’” says Giles Parkinson, global equities fund manager at Aviva Investors. “Investors have to find ways of dealing with uncertainty. A little more diversification – without diluting into ignorance – can be helpful to protect against unknown unknowns.”
Prepare for turbulence
As Parkinson points out, a well-diversified portfolio is not just a random collection of assets, but a set of informed ideas about corporate and economic trends, some of which will challenge the prevailing market consensus. The art of portfolio management is about ensuring these cohere, such that the associated risks are not concentrated in a single geography, sector or factor.
The first step in this process is to ensure risks are properly monitored, accounted for and financially compensated. Under conditions of uncertainty, it is especially important to distinguish between risks taken as essential components of an intended strategy and those that are unintended consequences of certain market bets, argues Mikhail Zverev, head of global equities at Aviva Investors.
“Consider a hypothetical equity portfolio that is long-US, short-Europe. The fund manager might have structured the portfolio in this way because they prefer the US to Europe from a macroeconomic perspective. But they need to be aware that there are other dynamics involved.
“For example, the US is more tech-heavy than Europe, where the market is more industrials-focused. So, the macroeconomic bet introduces a further lateral risk: the portfolio is not just long-US, but long-tech and short-industrials. That, in turn, brings other interest rate-risk and factor-risk implications,” Zverev adds.
As well as tracking “known-known” risks, it is important to closely track and manage correlations
As well as tracking these kinds of “known-known” risks, it is important to closely track and manage correlations – the degree to which asset prices move in relation to each other. Portfolios in which different securities are positively correlated will not be sufficiently diversified, even if those are spread across different asset classes and geographic markets.
“Diversification is a massive cornerstone in investing, but only works when there is a low correlation between ideas, to reduce the total amount of risk,” says Wei-Jin Tan, who monitors risk across Aviva Investors’ multi-strategy funds. “Correlations are not static, so it is important to continually conduct scenario analysis to get a sense of how they change in different environments.”
For instance, bonds and equities have historically been negatively correlated during periods of market calm, but this can change quickly during crises. Tan’s team undertakes scenario analysis to determine how correlations alter under different pressures, using historic events as a guide, including the global financial crisis, Russian currency crisis of 1998 and the “taper tantrum” of 2013. This work informs forward-thinking risk management in the context of the portfolio as a whole.
One important measure in tracking correlations is their “unusualness”. Derived from the work of Indian scientist P.C. Mahalanobis – who developed the concept while studying the distribution of human skull sizes – unusualness is a statistical indication of turbulence that shows how correlations move in extreme situations.
“What we saw going into COVID-19 was that unusualness spiked massively, and that indicated the traditional equity-bond correlation may not hold,” says Tan. “Portfolio managers must constantly evaluate and adjust to ensure that the investments they have are the right ones, and that their risks are properly diversified.”
Resilience and stock selection
Risk management in times of uncertainty also requires a more granular analysis of the resilience of individual assets within portfolios, from the creditworthiness of a sovereign-debt issuer to the health of a corporate balance sheet.
The coronavirus crisis has provided a stress test for many business models. Companies with low leverage, comfortable cash buffers and a diversified range of customers and suppliers have generally proven to be more resilient than those that favour lean, cost-cutting efficiencies.14
Choosing a good company is about more than reading balance sheets and P&L statements, however. One crucial lesson to be learned from the events of 2020 is that organisations need to play close attention to the wider market and social context in which they operate.
Companies with strong ESG credentials are proving to be more resilient to the disruption
Companies with strong environmental, social and governance (ESG) credentials are proving to be more resilient to the disruption, perhaps because these firms tend to take a more careful and holistic view of their operations and those of their commercial partners. Over the longer term, they should also be better placed to cope with the biggest threat of all: climate change.
“Leaders in ESG are focusing on the resilience and sustainability of their business models,” says Jaime Ramos Martin, global equities fund manager at Aviva Investors. “Take supply-chain management: in order to be a leader in ESG, companies would have needed to better understand the carbon footprints and labour practices of their suppliers, which will have prepared them for the disruption when COVID-19 hit.”
ESG-focused companies also tend to fare better when it comes to public opinion. In Uncharted, Heffernan discusses what organisations can learn from the success of “cathedral projects” – long-term collaborative initiatives that depend on continual buy-in from a range of stakeholders.
Consider the European Organization for Nuclear Research (CERN), the multi-disciplinary scientific centre that has yielded world-changing discoveries (the confirmation of the Higgs boson theory) and technological innovations (the World Wide Web). Through shrewd governance and public communication strategies, CERN has been able to maintain its funding model for decades. Such projects demonstrate the relationship between long-term resilience and public legitimacy, Heffernan argues.15
“No organisation in the world can function without society,” she says. “We need educated people; we need roads and energy and light. The rule of law. Health. Clean air. These sorts of things are not optional extras.
“Every corporation exists within an ecosystem, and the corporation can only be as resilient as the society it inhabits. The health of the organisation depends on the health of the ecosystem, and the health of the ecosystem depends on the health of each individual company,” adds Heffernan.
In a market context, companies with strong ESG credentials are more likely to earn legitimacy through a sense of corporate purpose, which should help them come through crises such as COVID-19 and prosper over the longer term.
Aggressive tax avoidance, poor labour standards and community relations and a substandard environmental record will be harder to defend in a world that has suffered the collective hardship of the coronavirus. Companies that have demonstrated they are willing to do the right thing are more likely to retain the loyalty of their staff and customers. Firms that don’t do the right thing will also find it more difficult to access government bailouts when required, as Ramos Martin points out.
Qualitative metrics are more difficult to assess and therefore more rewarding for investors willing to do thorough due diligence
It is important for investors to keep track of these qualitative measures – and take steps to improve them through engagement with company management teams – as they assess the resilience of their portfolios. Such factors are important determinants of value over the longer term. Whereas quantitative factors can be plugged into a Bloomberg terminal and tend to be quickly arbitraged away, qualitative metrics like corporate behaviour are more difficult to assess, and therefore more rewarding for investors willing to do thorough due diligence.
Invest on the side of change
Perhaps the most important criterion for resilience in an age of uncertainty is the capacity to adapt when circumstances shift. Organisations that can trim their sails and adjust course when the weather turns are more likely to prosper than those that simply batten down the hatches. A recent study tracked how 1,000 publicly traded companies fared during successive crises: it found an ability to adapt to new conditions was a hallmark of the best-performing firms.16
Consider the example of Finnish electronics firm Nokia, once the world’s leading manufacturer of mobile phones. After the launch of Apple’s iPhone in 2007, and the advent of the Android operating system the following year, Nokia’s devices looked outmoded and its market share shrank. In 2009, McKinsey placed Nokia in the bottom 25th percentile of its ranking of global companies and predicted it would cease trading within two years.17
Research shows organisations often become more rigid and inflexible during crises, doubling-down on a single plan of action – but Nokia was different. Its management undertook wide-ranging scenario planning and encouraged teams to collaborate across departmental silos to develop a new vision for the company. As a result of these methods, Nokia successfully reinvented itself as one of world’s leading telecommunications infrastructure firms.
Similar principles can be applied in asset management. Teams that work across asset-class silos tend to be better able to spot underlying vulnerabilities and respond in a timely fashion when the market environment shifts. This may be because cross-disciplinary teams are better placed to pick up on informational signals that indicate danger ahead – those early signs of “dragon king”-style events.18
“At the beginning of this crisis, many companies were rushing to tap the credit markets to build liquidity; company managements were not speaking to shareholders as much during this ‘firefighting’ period,” says Zverev.
“By collaborating with our colleagues in the credit team, we were able to get a sense of how COVID-19 was affecting those businesses. This sort of connected thinking is always useful, but particularly so during a crisis.”
Collaborating across teams can also provide an indication of how new developments are likely to affect different industries and sectors into the future, and where fresh opportunities might arise. Rather than relying on the misleading certainties offered by algorithms or all-encompassing probabilistic risk models, this approach is based on patiently piecing together a holistic picture of the market using a range of perspectives.
Collaborating across teams can provide an indication of how new developments are likely to affect different industries and sectors into the future
Assessing the impact of new developments is as much art as it is science. As Munro argues, human investors are better placed than quantitative models to make judgements based on informational inputs, not just market outputs, and to spot the inflection points between different regimes.
“Quant models work on the assumption the past is the best predictor we have of the future. But while you can take lessons from something that’s worked well over the last 20 years, that might have been because interest rates were going down over that period. How is the situation going to change when interest rates rise? What we’re always trying to do is to identify the impact of new trends – that’s where humans can offer value over machines,” Munro says.
When the information changes…
So, what changes COVID-19 will bring? A survey of the crisis so far suggests some potential future scenarios.
Technology giants, already among the world’s dominant companies, could grow even stronger amid rising demand for networking software. The pressures on global supply chains may prompt a shift from lean, “just-in-time” efficiency to “just-in-case” disaster planning. The car industry, which looked to be challenged by government decarbonisation measures before the crisis, could make a comeback as city dwellers grow wary of using busy public transport.19
Flexibility and humility are the hallmarks of longer-term success in asset management and beyond
These kinds of theses about the future can help teams develop what Kay calls “reference narratives”, guiding investment decisions. But in unpredictable times, these narratives need to be open to challenge and revision when the picture changes. Flexibility and humility – not unshakeable certainty in the wisdom of one’s own decisions – are the hallmarks of longer-term success in asset management and beyond.
No-one knew this better than John Maynard Keynes. In the years after he outlined the distinction between risk and uncertainty, Keynes managed an investment portfolio on behalf of King’s College, Cambridge. He grew the value of the fund with a series of accurate market bets based on his forecasts about the business cycle – but he did not anticipate the Wall Street Crash of 1929, which damaged the fund’s value (and cost him 80 per cent of his personal net worth).20
Other market forecasters of the era were also caught out: Irving Fisher, then the world’s most famous economist, claimed stocks had reached “a permanently high plateau” only days before the crash. But unlike Fisher, who doubled down on bad market bets, Keynes was willing to change his mind. This enabled him to adapt to the shifting post-Crash environment and cope with the adverse consequences of his mistakes. As Keynes famously put it in 1940, after he had adjusted his investment strategy and recouped most of his losses: “When my information changes, I alter my conclusions.”
At a time of radical uncertainty, investors everywhere would do well to heed his advice.