It is widely recognised that economic losses due to natural disasters have been increasing exponentially in the last decades. The main drivers of this trend are the increase in population and the growth in wealth per capita. With more and richer people, it is not surprising to find an increase in disaster losses. More surprising is the fact that, in spite of growing investments in risk reduction, the growth in losses has been as fast as economic growth (eg Miller et al 2008), or even faster than economic growth (eg Bouwer et al 2007). Anthropogenic climate change does not seem to play a significant role in these evolutions (Bouwer 2011). In the US, the trend in hurricane losses relative to wealth can be almost completely explained by the fact that people take more and more risks, by moving and investing more and more in at-risk areas (Pielke et al 2008). What are the reasons behind this trend? This column proposes an alternative to the ‘usual suspects’.
Most of the time, the explanations offered for this increasing risk-taking trend focus on market failures and behavioural biases. There are transaction costs: since the information on natural hazards is not always easily available, households and businesses may decide not to spend the time, money, and effort to collect them (Hogarth and Kunreuther 1995). There are also externalities and moral hazards: since insurance or post-disaster support are often available (especially in developed countries), households and firms do not pay the full cost of the risk, and may take more risk than what is socially optimal (eg Laffont 1995). Irrational behaviours and biased risk perceptions also play a role: individuals defer choosing between ambiguous choices (Tversky and Shafir 1992); they have trouble taking into account events that have never occurred before (Tversky and Kahneman 1974); and they do not always take long-term consequences into account (Michel-Kerjan 2008, Kunreuther et al 1978).
There is no doubt these factors play a role. But in recent research (Hallegatte 2011), I suggest that this move toward at-risk areas could also be a rational decision – motivated by higher productivity in at-risk areas – rather than a market failure.
Investing in risk
My analysis is based on a simple model where investments are assumed to be more productive in flood-prone areas, because there are close to coast and rivers, lowering transportation costs, or because of urbanisation, which leads to concentration and development in river flood plains. Economic agents are then assumed to decide rationally on (i) how much to invest in flood protection; (ii) how much to invest in flood-prone (protected) areas; and (iii) how much to invest in safe areas. The main mechanism of the model is really simple: with economic growth, we have more resources to dedicate to disaster protection; but with better protections, disaster probability is reduced, creating an incentive to increase investment in more-productive risky areas. So when a protection fails or is overtopped by an exceptional events, consequences are larger.
As a result, the model predicts that disaster frequency decreases with economic growth, but that consequences are larger when a disaster occurs. The share of capital and population installed in flood-prone area can be increasing with growth, making disaster consequences (when a disaster occurs) grow more rapidly than wealth. With economic development, therefore, we are moving toward a world of fewer but bigger disasters, even relative to the size of the economy.
Figure 1. How development increases vulnerability to the most extreme events.
And, depending on model parameters, the average annual losses can – counterintuitively – grow more rapidly than wealth and income. It means that in a world of rational decisions, it can make sense to behave in a way that makes average disaster losses increase, even in relative terms. There is thus no reason to hypothesise that more development means less vulnerability to disasters.
And this is assuming that all economic actors are rational. If one accounts for behavioural biases – especially the tendency to forget about risk when no event occurs during a long period of time – then the model suggests that even larger disasters are to be expected. With reduced disaster probability, the time lag between events increases, worsening this memory problem.
These results are consistent with what we observe. They are in line with observations from the UN International Strategy for Disaster Reduction (2009), which finds that poor countries suffer from frequent and low-cost events, while rich countries suffer from rare but high-cost events. And even if rich countries are on average less vulnerable than poor ones, the trends in each country are not what one would expect in a world where economic growth automatically reduces vulnerability.
The main mechanism is well illustrated by the case of Japan. Thanks to risk-management practices, Japan can cope with earthquakes (and tsunamis) that would cause catastrophes in other countries. But this resilience allows for higher investments in at-risk areas – such as coastal zones – and exceptional quakes like the recent Tohoku Pacific earthquake can then lead to immense losses.
What does it mean for development and risk-management policies?
First, disasters are likely to become rarer, but also larger. Crisis management tools will thus become increasingly important, and new tools will be necessary, including government disaster reserve funds (such as the FONDEN in Mexico) and multi-country disaster-risk pools (such as the Caribbean catastrophe risk insurance facility or the European Union Solidarity Fund). In general, international solidarity is likely to become even more important than it is today, and everybody would gain if it were organised in a more rational way.
Second, economic growth has no theoretical reason to reduce disaster losses, even in relative terms; reducing disaster losses requires specific, targeted policy actions. But the typical approach for flood mitigation is based on ‘zoning’, ie on the definition of flood-prone areas where investments are prohibited. This approach faces difficulties in implementation and enforcement. A more flexible approach that accounts for the benefits from investment in at-risk areas – through economic analysis and consultative processes – would be more effective. Policies should not systematically aim at reducing the level of risk. Instead, they should manage the level of risk, to limit disaster losses while making sure that we can still take the worthwhile risks that yield large benefits. Disaster risk management should be favoured over disaster risk reduction.
If a risk-increasing action is due to irrational behaviours, externalities, and imperfect information, then communication and land-use regulations are easy to implement and should be effective. But if this action is justified by the benefits derived from investing in at-risk areas, then risk reduction policies will fail or be economically detrimental, unless they provide alternatives to get similar benefits without increasing risk. If newcomers settle in risky areas of mega-cities because it is the only way to have access to jobs and opportunities, for instance, then prohibiting settlements in these areas is not a solution. Such a policy would face strong political opposition and might lead to informal settlements, thereby increasing the level of risk instead of reducing it. An efficient policy should instead propose viable alternatives to newcomers, for instance by providing cheap and rapid public transportation.
And more generally, disaster risk–management policies need to shift from a purely negative stance – indicating where it is prohibited to invest – to a positive approach – indicating where investments should be directed, and providing complementary measures that can make these investments as beneficial as those in at-risk areas. To do so, risk management should not focus only on at-risk areas, but follow a more holistic approach, one that is integrated in development planning. Sometimes, building a transportation infrastructure to connect job centres to safe locations is a more efficient way to reduce risks than building dikes.
Bouwer, LM (2011), “Have disaster losses increased due to anthropogenic climate change?”, Bulletin of the American Meteorological Society, 92:39–46.
Bouwer, LM, RP Crompton, E Faust, P Höppe, and RA Pielke Jr. (2007), “Confronting disaster losses”, Science, 318:753+.
Hogarth, R, and H Kunreuther (1995), “Decision Making Under Ignorance: Arguing with Yourself”, Journal of Risk and Uncertainty, 10:15–36.
Kunreuther, H, R Ginsberg, L Miller, P Sagi, P Slovic, B Borkan, and N Katz (1978), Disaster Insurance Protection: Public Policy Lessons, John Wiley and Sons.
Laffont, JJ (1995), “Regulation, moral hazard and insurance of environmental risks”, Journal of Public Economics 58(3):319–36.
Michel-Kerjan, E (2008), “Disasters and public policy: Can market lessons help address government failures”, Proceedings of the 99th National Tax Association conference, Boston.
Miller S, R Muir-Wood, A Boissonnade (2008), “An exploration of trends in normalized weather-related catastrophe losses” in Diaz, HF and RJ Murnane (eds.), Climate extremes and society, Cambridge University Press, 225–347.
Pielke, RA, Jr, J Gratz, CW Landsea, D Collins, MA Saunders, and R Musulin (2008), “Normalized hurricane damages in the United States: 1900–2005”, Natural Hazards Review, 9(1):29–42.
UN-ISDR (2009), “Risk and poverty in a changing climate: Invest today for a safer tomorrow”, United Nations International Strategy for Natural Disaster Reduction Global Assessment Rep. on Disaster Risk Reduction, 207pp.
Tversky A and D Kahneman (1974), “Judgment under Uncertainty: Heuristics and Biases”, Science, 185(4157):1124–31.
Tversky, A and E Shafir (1992), “Choice under conflict: the dynamics of deferred decision”, Psychological Science, 3(6):358–61.