Fixing the Hole When the Rains Come In


madelyn antoncic is a former vice president and treasurer of the World Bank.

Illustrations by Timothy Cook

Published July 31, 2018


According to Munich Re, the number of natural catastrophic events worldwide almost tripled between 1980 and 2017. The giant German insurer, which specializes in sharing the liabilities of other insurers, concluded that the trend may have as much to do with improved tools for tracking and reporting such events as with natural change. That may be so, but one thing is indisputable: the losses claimed increased seven-fold, from about $50 billion in 1980 to $354 billion in 2011, its peak year (think Japan’s tsunami). Last year, the second costliest year, losses amounted to $330 billion. That latter figure, incidentally was double the 10-year average, even after adjustment for inflation.

Yet, less than 30 percent of the losses in the past decade were covered by insurance. Households, businesses and governments absorbed the other 70 percent, creating monumental dislocation that tore apart lives and magnified the damage.

Lenders and investors share much of the risk. Indeed, some banks are starting to perform stress tests and scenario analyses on the impact that one of the most important potential sources of loss — climate change — has on their loan portfolios. Regulators are edging in the same direction; the Financial Stability Board, the international watchdog that advises member countries on financial matters, issued a report on climate-induced risk in June 2017 that recommended banks disclose stress results in their annual regulatory filings.

While recognizing the potential impact that natural disaster risk poses and using stress tests to assess financial institutions’ exposure is a good start, this effort will inevitably fall short of answering some fundamental questions about stability. The global economy is so interconnected and the secondary shocks from initial events potentially so large, that it is virtually impossible to measure the likely impact of a major natural catastrophe on a big bank’s portfolio. And, in any case, assessing and reporting risk is just a first step. Good risk management requires follow-through, mitigating and hedging risk.

Of course, private financial institutions themselves have a big stake in getting this right. But the risks of natural catastrophe also affect the stability of the global financial system. And thus public policy should aim to give private actors the appropriate incentives to hedge the residual risk in other ways.

Who’s at Risk

The costs of natural disasters are multidimensional. First, of course, are the costs of losses in life and limb, along with the costs of rescue and prevention of post-disaster losses from exposure, hunger and disease. Then, there are the direct property losses — the damage to roads, public buildings, houses, vehicles and the like. But on top of straightforward localized costs, economic losses filter through the global financial system, and the enterprises that constitute it, in myriad ways.

Consider, for example, that the closure or isolation of a key partner in a global supply chain may lead to great losses downstream. We saw this with the 2011 floods in Thailand. Not only did that country lose 5 percent of its GDP, supply chain disruptions of the same order of magnitude reduced industrial production in the Japanese economy by almost 3 percent in October and November of that year. Moreover, while Toyota’s plants in Thailand were not physically damaged by the floods, the company had to suspend production because of difficulties in obtaining parts from suppliers that had been damaged. And this disruption affected production lines in, among other countries, the United States and Canada. Similarly, the production of iPhones at multiple international sites was disrupted by the 2016 earthquake in Japan that shuttered the Sony plant that made the image sensors used in the devices.

Assessing and reporting risk is just a first step. Good risk management requires follow-through, mitigating and hedging risk.

Even in the absence of disruption to corporate facilities, business interruption can occur in the wake of the loss of public infrastructure or, for that matter, because of employee trauma. This was the case in Turkey after the Marmara earthquake in 1999 that killed 18,000 people, where businesses could not restart operations for an average of 35 days even though they did not incur direct damages.

Likewise, even an event in one country that has yet to occur can have significant knock-on effects around the world, showing up in rising costs for agriculture and end products due to the expected impact on the availability of key commodities. For example, a few years ago the price of nickel spiked on fears that a looming El Niño-induced drought in nickel-rich Indonesia would cut the availability of hydropower to operate the nickel smelters.

Think Globally, Act Nationally

Just as there are multiple avenues by which the costs of natural disasters diffuse through the global financial system, there are multiple drivers behind the growth in economic costs over time. These include population expansion and urbanization (often without the benefit of effective building codes), as well as climate change and seemingly unrelated environmental degradation. To the latter point: the wholesale paving of greater Houston reduced the land able to absorb heavy rains, leaving structures far more vulnerable to flooding from Hurricane Harvey in 2017.

Initiatives to prevent or mitigate these losses have to begin at the country level. Left unaddressed, governments are exposed to substantial fiscal instability — not only in the immediate aftermath, but also in the longer term because of lost employment, business disruptions and lower tax revenue.

While some risks cannot be eliminated, the economic costs of the disasters that do happen can be reduced through better planning, tougher building codes that are strictly enforced and early warning systems. Some countries have pushed back at the natural tendency of politicians to kick the proverbial can down the road and are beginning to integrate disaster risk management into development strategies and fiscal planning. The United Nations and development-finance institutions, including the World Bank, are nudging planners in this direction, offering technical expertise to facilitate the process.

New tools offer insight into which risks to transfer, mitigate or hedge. Yet even though the tools are increasingly available, serious barriers remain to making cost-effective use of insurance.

New tools allow countries to assess which risks to absorb, such as those with a high probability but a relatively low cost (think floods), and to use reserves or contingent lines of credit to cope with those risks. By the same token, they offer insight into which risks to transfer, mitigate or hedge — that is, those with a low probability of occurrence but a very high potential cost, such as earthquakes.

Yet, even though the tools are increasingly available, serious barriers remain to making cost-effective use of insurance — especially in the public sector in emerging economies. Relatively few of these countries have the expertise and institutional framework to evaluate risk, prepare a disaster-financing strategy and then price and negotiate market-based risk transfers. What’s more, the more immediate requirements of government compete for scarce fiscal resources, and the overriding importance of insurance only becomes apparent after a major disaster. In fact, 95 percent of financing for disaster since 1980 has been for after-the-fact emergency response.

This is not to suggest there aren’t exceptions. In fact, some emerging-market countries have employed very sophisticated hedging strategies. At the World Bank, we executed the largest weather and energy derivative the market has ever seen to help a country that is hydropower-dependent mitigate the risk of droughts and high energy prices. For its part, Mexico has for years made use of hedging, both to protect its oil revenue from significant price swings and to mitigate the cost of natural disasters such as hurricanes and earthquakes.

Since natural disasters don’t respect national borders, Mexico extended its coverage in 2017 to hedge risks in neighboring states in the United States. A big event there could damage pipelines, reducing sales of Mexican fuel. In fact, this coverage went much further. Had the winds in Harvey reached a designated threshold, the insurance (here, in the form of a “catastrophe bond”) would have triggered payments shielding the loss in remittances from Mexicans living in the damaged territory of Texas.

Note, by the way, that the net cost of such insurance is lower than it might first appear. Hedging exposures have led to ratings upgrades for some countries, with the obvious knock-on effect of reducing the interest paid on government debt.

While my focus here has been on lower-income countries, the rich industrial nations also fall short on strategies for risk prevention and mitigation, as well as insurance penetration. In the United States, losses in 2017 represented half of total global losses from natural disasters, significantly higher than the long-term average of slightly over 30 percent. Munich Re noted that better mitigation — for example, stronger building codes in Florida — could have reduced those in a cost-effective manner. And even though the tools, expertise and institutional frameworks for risk management are widely available in the United States, only about 40 percent of the 2017 losses were insured. Plainly, much more needs to be done to reduce the costs of natural disasters when they happen and to engage the broad array of hedging instruments available to buffer the direct and indirect costs.

Corporations, Step Up

Economic development in general and infrastructure investment in particular mean the accumulation of assets, a good part of which will be in areas vulnerable to natural disaster. As important, those vulnerabilities are growing, thanks to poorly managed urbanization on the one hand and climate-induced weather severity on the other.

According to the IMF, emerging-market and developing economies accounted for more than three-quarters of the global growth in output (measured in terms of purchasing power) last year. And according to the OECD, the stock of foreign direct investment in these economies approached $10 trillion by the end of 2016 — up five-fold since 2000. This is on top of committed investments from the development institutions.

Yet there are only limited opportunities for transferring natural disaster risks borne by private investors in these countries to financial markets in the advanced industrialized world, where risk can be managed at much lower cost. This is a looming market inefficiency that generates important residual risks to investors and creditors as well as contingent liabilities for governments.

Managing catastrophic risks at the corporate level in developing countries is important not only for the firms directly exposed, but also for banks, pension funds, insurance companies and other institutional investors as they increase their involvement in infrastructure financing. Nonetheless, scarcely half of non-financial firms report having a formal enterprise risk management framework of any kind — never mind one that incorporates catastrophic risk. Indeed, risks arising from uninsured infrastructure investments may well become an important mechanism through which risks are magnified and spread across the globe.

Incentives, Incentives, Incentives

As noted, rating agencies have incorporated disaster risk management hedging by countries into their ratings analyses and have upgraded some to reward them for contingent liability planning. Financial regulators could also help to create incentives for disaster risk management by taking such contingency liability planning into consideration, effectively rewarding banks (and their customers) by lowering the risk weights on national governments that hedge.

This is no small matter, since for some small countries a natural disaster can cost up to 100 percent of its GDP. However, losses can be significant even in developed countries. Japan, for example, lost over $300 billion (4 percent of its GDP) from damage due to the magnitude-9 earthquake and tsunami in 2011.

Likewise, regulators can ease capital requirements for banks that hold the paper of enterprises choosing to hedge their risk. After all, by definition these sovereigns and corporations are less risky, and so should be given the recognition through capital relief.

Reinsurers Can’t be the Only Game in Town

Given the magnitude of the global costs of natural disasters and the size of the gap in insurance coverage, the much-needed increase in hedging would probably outstrip reinsurance carriers’ capacities. While in recent years reinsurers have stepped up by substantially increasing their capital bases to support catastrophic risk coverage, their resources simply wouldn’t be adequate in all conceivable crises. Financial markets need to create new insurance products to fill the breach.

One relevant product mentioned earlier, cat bonds (aka catastrophe bonds), first appeared in the early 2000s. But they never developed into an asset class with deep, liquid markets. By contrast, the market for insurance-linked securities has emerged as a solution that can compete with negotiated reinsurance. In the latter, risks are hedged mostly through bilateral swaps, which are typically 12 to 18 months in tenure. The advantages of ILS over hedging via reinsurance swaps are clear. ILS are priced in transparent markets. They cover longer periods, sparing the need to renegotiate the terms frequently. And they are transferrable: sponsors can hedge any type of peril in the market, as we did at the World Bank when we created a capital market note that provided for rapid response coverage for pandemics — including Ebola.

The market for insurance-linked securities does double duty, driving down the cost of traditional reinsurance by expanding the base of investors in hedging securities. Indeed, ILS are especially appealing, both because the return on them is totally uncorrelated with other traded assets and because investors continue to reach for yield in a low-interest rate environment. But the market does have a way to go. According to Aon, the big U.S. based insurance broker, ILS represent just 14 percent of total reinsurance capital.

Time for Action

It’s easy to focus attention on natural disasters in their immediate aftermath. What’s hard is getting policymakers and the public to plan for disasters that haven’t happened — and may never. But we can be fairly sure that, overall, the number and cost of such catastrophic events are growing, and with it the penalties of myopia. Compounding the problem, the ever-increasing integration of the global economy is making it harder to duck the consequences of poor planning even when disaster strikes on another continent.

The good news is that the tools to manage risk more efficiently are within reach. What are needed are policymakers, regulators and C-suite planners with the foresight to imagine a more dangerous world and arrange their priorities accordingly.

main topic: Insurance
related topics: Climate Change