“By the end of the century, infectious diseases will no longer pose a significant threat to human beings, even in the poorest countries”. Sounds fantastical? Yet, this was a prediction that underlay an accord called “Health For All 2000” signed by UN members in 1978. Around 13 million people die every year from infectious diseases, primarily from cholera, AIDS, tuberculosis and malaria.
Optimistic end-of-century forecasts like the UN one were commonplace. Their fault was not necessarily poor forecasting methodology or collective delusion, but probably a serious underestimation of the risks inherent in the predictions they were making. Today, it appears that the long-term trend of progress in the control and eradication of infectious diseases that fuelled this optimism has been reversed, as a consequence of the spread of drug-resistant microbes, the emergence of new devastating infections and the surge in unhelpful socio-economic factors like megacities with poor sanitary conditions, cross-border migration and the like.
Risks have always been around, but significant changes have occurred in recent years in the nature of major risks, and further changes can be expected in the future, both in range and in our capacity to manage them. Important factors include extreme and volatile weather conditions; population explosions; over-concentration and polarisation of wealth and economic activity; not to mention risks associated with technological and scientific discovery. And of course, there are risks from conflict and global terrorism.
Disasters can have huge social, economic and political impacts, as can our reactions to such risks. Take the terrorist attacks in the United States on 11 September 2001. Apart from the 3,000 deaths they caused, lost physical assets were estimated in the national accounts to amount to over US$15 billion. Rescue and clean-up amounted to some $11 billion. Lower Manhattan lost approximately 30% of its office space and scores of businesses disappeared. Some 200,000 jobs were destroyed or displaced, at least temporarily. Since the attacks, defence and security costs have risen.
Terrorism is at the fore of many people’s minds now, but it should not obscure other important risks. New major hazards are emerging, some of which are characterised by extreme uncertainty, even irreversible harm. According to various measures, the damage caused throughout the world by catastrophic events, seen as the realisation of major risks, seems to have increased in recent decades. First, the observed number of natural disasters, including floods, storms and droughts, has risen dramatically since the beginning of the 1960s. In the past decade, such disasters have resulted annually in some 79,000 fatalities, and 200 million people affected.
Fortunately, these human costs are by and large smaller than in the first part of the century, thanks to progress in health care, urbanisation and construction, and emergency services. Financially, though, damage has grown exponentially and is concentrated in developed countries (especially if insured damage is considered). Hurricane Andrew, that hit the United States and the Bahamas in August 1992, had 38 fatalities and caused economic losses close to $30 billion in 2000 prices. By comparison, Tropical Cyclone Gorky killed 138,000 people in Bangladesh in April 1991, with only modest insured losses.
Recorded technological disasters such as explosions, fires and transportation accidents have also risen rapidly in the last 40 years. Their human cost has increased in parallel. Most are relatively small events, but there are very large accidents too, such as the 1987 ferry collision in the Philippines (4,375 victims), the 1984 chemical factory accident in Bhopal, India (3,000 victims), or the 1986 nuclear reactor meltdown in Chernobyl (31 immediate victims, 135,000 reported affected, $2.8 billion of economic losses).
It is developing countries that suffer most from infectious diseases. But OECD countries also are vulnerable to risks of pandemics, the resistance of microbes, and food-borne diseases. Even tuberculosis, which was almost eradicated from Europe, has bounced back in recent years.
An important feature of modern disasters is that they can rapidly spread in space and time. To quote a major reinsurance company, disasters can reach “a level that can result in the collapse of the economic system of entire countries and may be even capable of affecting financial markets throughout the world” (Munich Re, 2000).
Admittedly, data on impacts of disasters have to be taken with caution. Better tools of observation highlight them more. Also, costs are hard to compare. The 1997 floods in Poland appear as a medium-size disaster in absolute terms (some $3 billion), but still came to 3% of GDP. Insured losses cannot be used as reliable indicators of economic loss either: infrastructure, which often represents a large share of physical damage, is not always insured. [Nor, increasingly, are risks from terrorism.] And the notion of disasters does not capture various risks with diffuse effects, such as pollution or asbestos.
Dealing with risks
So, we know the future will bring significant risks, the question is how to deal with them. This is a question the OECD, notably through its International Futures Programme, is busily examining in a major project on Emerging Systemic Risks in the 21st Century. The final report is expected to be published in March 2003. The project identifies a number of key areas in risk management in need of improvement. For instance, assessment and prevention of emerging risks need to be designed in a more forward-looking way, and implemented with greater coherence. In order to reduce the vulnerability of complex systems (such as critical information infrastructures and global supply chains of essential products) more attention needs to be paid to diversifying sources and providing alternative back-up solutions.
For instance, of growing concern is the issue of risk inside our ever more complex systems and the potential for negative consequences of an accident. According to US scientist Charles Perrow, the evaluation of the consequences of an accident is often limited to short-term losses, while the long-term consequences, that are admittedly much harder to estimate, can be considerable, too. An accident often results from the coincidence of two or several failings, each of which had been anticipated by the designers and system operators, but the conjunction of which was totally unexpected. The trouble is, the probability of such a conjunction increases with the system’s degree of complexity. And a lack of prior knowledge can even lead to corrective action that will worsen the situation. In Mr Perrow’s view, accidents in complex systems are not only difficult to predict, but are effectively “embedded” in the system.
Still, forewarned is forearmed, and risk management is not only essential, it appears to pay off, having been adapted to complex systems like nuclear reactors. The tricky point there is that accidents are so rare that there is little empirical evidence for study. As a result, for nuclear reactors, risk assessment is often based on failure mode and effect analysis. A first step consists in identifying all the ways in which the system’s functioning can be altered, and finding all the possible chains of events (involving material or men) that could lead to such failure modes. Elementary failure probabilities are then estimated and rules of risk management are derived.
Nuclear risk management is about preventive action, though it involves a dose of the so-called precautionary principle, which consists of recognising the need for action against risks even in the absence of a clear assessment of their nature and magnitude. But whereas preventing a nuclear disaster implies acting against revealed and quantified risks, precaution implies acting against potential risks, in contexts often characterised by a high degree of scientific uncertainty and controversy. We might not really know if genetically modified food damages health or not, but we know what a nuclear accident might do. The notion of precaution has existed for a long time in food and health regulations. Numerous references to the precautionary principle can be found in international agreements, notably the 1992 Rio Declaration on Environment and Development and the 2000 Protocol on Biosafety. Various countries have integrated the principle into their national legislation and it has become a useful tool in international co-operation.
These are just some of the issues considered by the OECD International Futures Programme and more can be consulted in the documentation (see www.oecd.org/futurestudies). There are many unanswered questions which the forthcoming report, which draws on international expertise and experience, will endeavour to answer. But remember, no matter how many questions are resolved, there is always a risk something will go wrong.
For more on the forthcoming report, Emerging Systemic Risks in the 21st Century (preliminary title), e-mail email@example.com
Lenain, P., Bonturi, M., and Koen, V., (2002) “The fallout from terrorism: Security and the economy” in OECD Observer No. 231/232, May 2002; article available free at www.oecdobserver.org
OECD–NEA (2001), Methodologies for Assessing the Economic Consequences of Nuclear Reactor Accidents, Paris.
Perrow, C. (1984), Normal Accidents, Basic Books, New York.
©OECD Observer No 235, December 2002