Risk and reaction: dealing with interdependencies.
Kunreuther, Howard
With the increasing concentration of people and businesses in
high-risk areas and the increasing interdependencies within the world,
catastrophes are more likely than ever. Consider the disasters of
Hurricane Katrina and the September 11 attacks. Not only was Hurricane
Katrina the largest natural disaster in US history, but it also
demonstrated how economic and social activities affected by the event
translated into global risks with worldwide ripple effects. Katrina had
major impacts on several international markets, such as oil, gas, and
insurance. The terrorist attacks of September 11 had the immediate
impacts of property damage and fatalities, killing more than 3,000
people from more than 90 countries, injuring about 2,250, and inflicting
direct damage estimated at nearly US$80 billion. They also caused
significant business interruption losses throughout the world and
reshaped international relations.
[ILLUSTRATION OMITTED]
Natural disasters and terrorist attacks are examples of what have
been classified as low probability-high consequence (LP-HC) events.
Despite the first half of their title, these events are now in the
headlines with increasing frequency. There are features of these
potentially catastrophic events that need to be carefully examined in
order for individuals, firms, and governments to take steps today to
reduce the risk of their occurrence in the future and to reduce the
consequences should they occur. Every government in the world faces the
question of how to help its citizens face the risks of catastrophe.
Reducing these risks requires a concerted effort by individuals and
firms. This poses a set of challenges due to the difficulty of
anticipating LP-HC events.
It Will Not Happen to Me
Prior to catastrophic events, decision-makers often assume that
these disasters will not happen to them. Hence they have little interest
in undertaking costly protective measures. Only after a disaster is
there concern with taking steps to prevent another such catastrophe from
occurring. Hurricane Katrina and the September 11 terrorist attacks
highlight this point.
Before Hurricane Katrina, the US Army Corps of Engineers did not
adequately shore up the existing levees to protect the New Orleans
metropolitan area from flooding. The reasons were several: cost
increases, design changes caused by technical issues, environmental
concerns, legal challenges, and local opposition to portions of the
project. The Corps' project fact sheet from May 2005 noted that the
appropriated amounts in the President's budget for fiscal year 2005
were insufficient to fund new construction projects that included levee enlargement. The interested parties concerned with the hurricane threat
to the Gulf Coast acted as if a disaster such as Katrina would not occur
in the next few years. This is sometimes referred to as the NIMTOF--Not
in My Term of Office--phenomenon. It was convenient to ignore the
numerous alerting studies--which stated that the existing levees were
inadequate and that one of the nation's highest priorities should
be to reinforce these flood control projects--or even the October 2004
National Geographic warning about what could be the most devastating natural disaster in US history. The scenario became reality 10 months
later.
The federal government is now committed to providing disaster
assistance to the victims of Katrina. There is serious discussion of
reinforcing and rebuilding the levees in New Orleans and other parts of
the Gulf Coast. A few days after Katrina made landfall, the US Senate
needed just a few hours of discussion to vote to grant nearly US$60
billion in federal aid. As of February 2006, six months after Hurricane
Katrina, some US$88 billion in federal aid had been allocated for
relief, recovery, and rebuilding, with another US$20 billion requested
to help victims of the storm and the region recover and rebuild.
Likewise, prior to the September 11 attacks, insurers viewed losses
from terrorism as so improbable that they never explicitly considered
the risk when pricing their standard commercial insurance policies.
Losses from terrorism were never excluded from so-called
"all-risk" policies, with the exception of some marine cargo,
aviation, and political risk policies. Even the first attack on the
World Trade Center (WTC) in 1993 and the Oklahoma City bombing of 1995
were not seen as threatening enough for insurers to consider revising
their view of terrorism as not worth considering when pricing a
commercial insurance policy.
[ILLUSTRATION OMITTED]
The attacks of September 11 produced a fundamental change in how
terrorism is perceived in the United States. Insurers and re-insurers,
who paid the bulk of the US$35 billion of insured losses from the
terrorist attacks, were reluctant to continue offering protection except
at very high prices. As a result, insurers excluded terrorism damages
from their "all-risk" commercial policies when renewing these
policies. Firms demanding insurance protection against such losses were
forced to purchase a policy that included terrorism as a specific cause.
They often had difficulty finding an insurer offering such coverage at a
premium they were willing to pay. Sometimes they could not find a seller
willing to provide terrorism insurance at any price.
The lack of availability of terrorism insurance soon after the
September 11 attacks led some private sector groups, such as the
construction and real estate industries, to call for federal
intervention. In response, the US Congress passed the Terrorism Risk
Insurance Act of 2002 (TRIA) at the end of 2002 and renewed the act for
two years with only minor changes at the end of 2005. Under TRIA,
insurers are now obligated to offer terrorism coverage to all of their
commercial policyholders. Firms are not required to purchase this
insurance unless mandated by state law, as is the case with
workers' compensation protection. To date, more than 50 percent of
commercial enterprises in the United States have bought terrorism
insurance. It is very likely that this proportion would increase
significantly should another attack occur on US soil.
Interdependent Security
An interdependent security (IDS) setting is one in which each
individual or firm that is part of an interconnected system must decide
independently whether or not to adopt protective strategies. These
measures can reduce the risk of a direct loss, but there is still some
chance that the individual or firm can suffer damage from others who do
not adopt similar measures. The economic incentive of any decision-maker
to invest in protective actions thus depends on whether it expects
others to follow suit--a classic case of game theory. The protective
strategies can be direct risk-reducing measures as well as
information-gathering and preparedness strategies. That such events are
typically probabilistic and that risk is often determined in part by the
behavior of others gives a complex structure to the incentives that
individuals or firms face when deciding whether to invest in risk
mitigation measures.
For many IDS problems, if an individual or firm thinks that others
will not invest in security, it has less incentive to do so if such
security measures cannot provide protection against ripple effects from
the failures of others. On the other hand, if each decision-maker
believes that others will invest in security, then its optimal strategy
is also to undertake protective measures. So there may be equilibrium
solutions wherein no one invests in protection, even though all would be
better off if each of them had incurred this cost. Airline baggage
screening is an IDS scenario that illustrates this problem.
An airline has to determine whether it wants to invest in baggage
screening security, knowing that even if it takes this action it may
face a security risk from a dangerous bag loaded onto its plane by
another airline. The airline faces this risk unless it inspects all
transferred bags. Lest this point be considered only theoretical, recall
the crash of Pan Am 103 in 1988. Terrorists checked a bag containing a
bomb in Malta on Malta Airlines, which had minimal security procedures.
The bag was transferred in Frankfurt to a Pan Am feeder line and then
loaded onto Pan Am 103 in London's Heathrow Airport. The bomb was
designed to explode above 28,000 feet, a height normally first attained
on this route over the Atlantic Ocean. The terrorists deliberately
exploited the widely varying security procedures across the airlines.
This problem is common to other transportation modes, where there are
interconnections between transfer points in the network.
In the wake of September 11 and Hurricane Katrina, the private and
public sectors share an interest in making social and economic systems
less vulnerable to disasters. There is a growing interest in protecting
the critical infrastructure--such as transportation, telecommunication,
electricity, and financial services--to assure the social and economic
continuity of the nation. One challenge is the existence of
interdependent operations between multiple infrastructures in different
sectors. For example, financial systems or emergency services are highly
dependent on telecommunication operations, which are highly dependent on
electricity. When the interdependencies cut across sectors, the nature
of the risks is often not well understood. These risks pose special
policy challenges at the national level. The problem has even more
economic significance and presents greater challenges for coordination
between countries should there be interdependent effects at the
international level.
Risk Management Challenges
The combination of an "it will not happen to me" attitude
and the interdependent security nature of many events with catastrophic
potential calls for a rethinking of strategies for managing these risks.
When decision-makers decide not to pay attention to these risks prior to
the occurrence of an event and the public sector then responds with
large-scale disaster assistance or recovery aid, we need to examine
whether there are better ways of dealing with this problem. In the case
of natural disasters, we should ask: What is the responsibility of
citizens residing in hazard-prone regions to protect themselves against
the consequences of these events prior to their occurrence? When it
comes to terrorism, we can ask businesses a similar question: What will
it take for them to purchase insurance to cover losses from damage and
business interruption before the next terrorist attack occurs?
The nature of interdependencies poses additional challenges for
risk management. An airline will have less economic incentive to invest
in risk-reducing measures if it knows other airlines will not follow
suit. This is a principal reason that new federal baggage and passenger
security measures were instituted in airports after September 11.
Government officials were concerned that no individual airline would
undertake these measures on its own. With respect to the impact of
catastrophic accidents on the survival of firms, what economic incentive
does any unit in a decentralized firm have to invest in protective
measures that adversely affect its balance sheet, if other divisions in
the organization are not taking similar actions? A culture of
risk-taking can spread throughout the firm. The knowledge that a few
groups are taking their chances lowers the incentives that others have
to manage their operations carefully.
In this context, the private market alone will not be able to
reduce future losses from catastrophic risks. There is little incentive
for individuals or firms to invest in protection if they are not
concerned about the consequences of such an event and know that they may
suffer losses from other unprotected firms even if they themselves have
invested in mitigation measures. What is particularly unfortunate about
this state of affairs is that if everyone took protective actions, they
would be better off in the long-run.
Managing Risk: Reducing Catastrophic Losses
Insurance can play a central role in reducing future losses and
providing protection. But this policy tool needs to be supplemented with
other actions that require some type of public sector involvement. In
theory insurance can encourage individuals to adopt loss reduction
measures by lowering premiums or deductibles. For this to occur,
however, rates have to reflect risk. Even then it is hard to sell this
idea to policyholders because the premium reduction given to the
property owner is normally relatively small compared to the cost of a
mitigation measure.
To illustrate, suppose that a resident or business can reduce its
potential wind damage caused by a hurricane by bracing its roof trusses
and installing straps or clips. These modifications are relatively
inexpensive and will result in the property owner paying a lower
insurance premium. Suppose the cost of this measure is US$1,500, but it
would prevent US$30,000 in damage if a hurricane hit. If the annual
likelihood of a hurricane in the area is estimated to be 1 in 100, then
the expected benefit from strengthening the roof would be US$300. A
risk-based insurance premium would thus be reduced by US$300 if the home
or business undertook this loss reduction measure. If the property owner
was myopic and only compared the cost of renovations with the expected
benefits in reduced premiums over the next several years, he would not
invest in mitigation. If the property owner planned to move in the near
future, the investment would be attractive only if it led to higher
property values. A property owner with budget constraints would have an
additional reason not to invest in this loss reduction measure.
The public sector can encourage the adoption of cost effective
mitigation measures, such as roof mitigation in hurricane-prone areas,
by developing building codes. Building codes are normally imposed on new
structures but not on existing ones. To encourage property owners to
retrofit their property built before the code was established, banks
should consider providing long-term mitigation loans that could be tied
to the structure rather than the owner. These loans would provide the
funds necessary to retrofit the building and would be economically
advantageous as long as their cost was less than the reduction a
homeowner would receive in insurance premiums. In the above example, if
the bank provided a 20-year home improvement loan at an annual interest
rate of 10 percent, then the annual loan payment would be US$145.
Because the reduction in the insurance premium would be US$300 for
strengthening the roof, the property owner would save US$155 by taking
out a loan to invest in this mitigation measure.
The challenge in implementing such a program is that the premiums
charged to those residing in the highest-risk areas are likely to be
considerably larger than they are today with current cross-subsidy among
different locations within a state or even between states. In fact, many
states regulate rates so that premiums do not reflect the actual risks.
Furthermore, some homes in high-risk areas are owned by low-income
families who cannot afford the costs of insurance or the costs of
reconstruction should their house suffer damage from a natural disaster.
Because uninsured, low-income victims are likely to receive federal
assistance after a disaster, loans or grants for insurance or mitigation
measures to this group would reduce the cost to all taxpayers following
a disaster and also reduce the uncertainty associated with the level of
reimbursement these families would receive in case of a disaster. A
risk-based insurance program with subsidies to low-income individuals in
the form of insurance vouchers, similar in concept to food stamps, would
enable insurers to set the appropriate rates over time.
[ILLUSTRATION OMITTED]
Where Do We Go From Here?
Misperceptions of risk and interdependencies pose enormous
challenges for individuals, firms, and the public sector in managing
catastrophic risks. By recognizing the nature of people's decision
processes and the limitations of the private sector in coping with these
risks on its own, creative partnerships between the public and private
sectors can be developed. There is the inevitable tension between
letting individuals make their own choices and imposing requirements on
them. But because individuals ignore a problem because they do not think
it will happen to them and believe the public sector will rescue them if
it does, there are grounds for imposing requirements in advance of a
possible catastrophe. This point was highlighted almost 30 years ago by
Finn Kydland and Edward Prescott in their Nobel-prize winning paper
arguing for rules rather than discretion. The challenges of managing the
risks of global catastrophe are even more significant today than they
were then.
The gravity of the challenges of interdependent risks is
particularly apparent when considering a potential avian influenza outbreak. Much as we were aware of the possibility of an event like
Hurricane Katrina, we are aware of the danger of a possible outbreak as
a result of human-to-human transmission of the virus. Each country in
the world can undertake its own preparations, but if some do not, the
scope and spread of the disease will be much greater than if every
country took the same measures to prevent an epidemic from occurring.
Countries must be induced through collective agreements to work together
to mitigate the potential catastrophic risk of a pending avian influenza
outbreak. Unlike the case of Hurricane Katrina, however, there is no
global government to come to the rescue.
We need to rethink ways of reducing the likelihood and consequences
of catastrophic risks. During the past five years, the United States has
been hit by a series of unprecedented major disasters, opening a new era
of large-scale catastrophes. Perhaps some decision-makers still prefer
not to think about the problems caused by these disasters because it
means that they can devote time to the short-term issues that demand
immediate attention. After the next disaster strikes, the consequences
will be much greater than they imagined. More than ever, governments,
firms, and individuals must take steps to address these risks before
they become the next catastrophes.
HOWARD KUNREUTHER is the Cecilia Yen Koo Professor of Decision
Sciences & Public Policy and Co-Director of the Risk Management and
Decision Processes Center at the Wharton School of the University of
Pennsylvania.
RELATED ARTICLE: PRISONER'S DILEMMA
The Risks of Interdependent Security
This scenario of Interdependent Security (IDS) illustrates the
dilemma faced by two airlines in deciding whether or not to invest in
baggage screening security. The ideal solution would be for both
airlines to invest. If only one airline invests, not all bags are
screened and security risk is relatively high. If both airlines assume
the other will not invest, neither will invest and reach a equilibrium
solution without screening costs but a high security risk.
[ILLUSTRATION OMITTED]
Airline B invests in baggage Airline B rejects baggage
security security
Airline A Both incur security costs Airline A incurs security
invests in Low security risk costs
baggage Relatively high security risk
security
Airline A Airline B incurs security Equilibrium solution:
rejects costs Both avoid costs
baggage Relatively high security High security risk
security risk
Howard Kunreuther
