A 6-10 dossier articles

C© Risk Management and Insurance Review, 2005, Vol. 8, No. 1, 141-150

THE COLUMBIA SPACE SHUTTLE TRAGEDY: THIRD-PARTY LIABILITY IMPLICATIONS FOR THE INSURANCE OF SPACE LOSSES Piotr Manikowski

ABSTRACT

Space flights are no longer rare events, but the commonplace is not necessarily safe. When disaster strikes, as in the Columbia Space Shuttle disaster of 2003, third parties as well as those directly involved are financially affected. This article considers how these issues are treated under international law. It also analyzes what products the insurance markets offer as protection against such third-party liabilities.

INTRODUCTION On February 1, 2003 the Columbia space shuttle, the oldest of a fleet of four, was destroyed during reentry into the earth’s atmosphere, causing the death of all seven crew. The total damage is estimated at about US$3 billion. During the International Space Insurance Conference that took place in Florence (April 3–4, 2003), Paul Pastorek, General Counsel of U.S. space agency NASA reported the latest findings of the investigations into the loss of the Columbia space shuttle (Stahler, 2003). NASA had recovered 45,000 pieces of wreckage from an area 100 miles long and 10 miles wide. The material recovered comprised in terms of weight almost half the lost shuttle. The initial suspicion was that one of the brittle ceramic tiles on the underside of the wing had been damaged during take-off, allowing heat to enter into the wheel chamber. A video tape was recovered, but this stopped transmitting shortly before the crew realized that there were problems with the re-entry. NASA subsequently recovered an instrument used on the shuttle to record a multitude of technical data during each flight. These data revealed that the build-up of heat inside the right wing came from the leading edge of the wing, which was made of an extremely hard and tough material. The initial ceramic-tile theory thus seemed to be disproved. However, the official report has yet to be released. Was Columbia the victim of a collision with space debris, of which thousands of items are now littering the earth’s orbital paths? It may never be established with absolute certainty what really happened

Piotr Manikowski is with the Poznań University of Economics, Insurance Department, al. Niepodleglosci 10, 60-967 Poznań, Poland (e-mail: piotr.manikowski@ae.poznan.pl). This ar- ticle was subject to anonymous peer review. The author wishes to thank Peter Birks for his language revision of the text.

141

142 RISK MANAGEMENT AND INSURANCE REVIEW

at a speed of 21,000 kilometers an hour in the upper layers of the atmosphere above Texas.

Debris from the space shuttle fell to the ground, but did not cause serious damage. However, it remains possible that space exploration could inflict harm on third parties on the ground. This could evoke the civil liability of the guilty party. It is possible to buy third-party liability insurance for space losses.

GENESIS OF SPACE (SATELLITE) INSURANCE Until the mid-1960s the insurance market was not interested in the space industry, since it had been focused on the military aims of the United States and the Soviet Union. The launching of the first artificial earth satellite on October 4, 1957 and the sending of the first man—Yuri Gagarin—into space on April 12, 1961, accelerated the development of the space industry—including its commercial arm. It became clear to the insurance industry that there would soon be a commercial space market available for exploitation.

Insurance for space activities has evolved over many years through the collaboration of aerospace clients, brokers, and the underwriting community worldwide. The goal of that work was to provide flexible forms of insurance for a volatile class of exposure, which was not yet quantified by loss data.

In the formative years of the space age, projects were uninsurable: launch vehicles were unreliable and most of the payloads were experimental—the risk was self-insured by governments and space agencies that financed the flights. The first company to devote its attention to the use of this new technology for commercial purposes and to show an interest in obtaining insurance protection was American Communication Satellite Corporation (ACSC), founded in 1962. On April 6, 1965 ACSC obtained the first space insurance policy to protect the first commercial geostationary communication satellite Early Bird (Intelsat I-F1). The policy covered only material damages to the satellite prior to lift-off (pre-launch insurance for US$3.5 million) and third-party liability insurance for US$ 5 million (Daouphars, 1999).

In time, and with increasing experience of insurers and the insured, the insurance market developed a wider scope of space insurance cover. There are currently three basic groups:

1. Property insurance: (pre-launch, launch, in-orbit insurance);

2. Third-party liability insurance;

3. Warranty insurance (loss of revenue, launch re-flight (risk) guarantee, incentive payments insurance).

The third group is supplementary to property cover. In this study only third-party li- ability insurance is taken into consideration. It should be emphasized that, since the early days of satellite insurance, little notice has been taken of the issues connected with liability for space damages.

RISK OF THIRD-PARTY LIABILITY FOR LOSSES MADE BY SPACE OBJECTS Space activity and the use of spacecraft entail the possibility of inflicting damage on third parties, for which the owner or the user of a satellite is usually responsible. In the event

THE COLUMBIA SPACE SHUT TLE TRAGEDY 143

of the explosion of a rocket only a few meters above the ground, the potential loss could be enormous.

In connection with the specificity of space activity and its “over-territorial” character, it was decided that the responsibility for damages should be regulated by international law. From the late 1960s a series of five treaties and conventions were agreed upon that covered the exploration of space and the legal ramifications for events on the ground:

� The Treaty on Principles Governing the Activities of States in the Exploration and Use of Outer Space, including the Moon and Other Celestial Bodies (the “Outer Space Treaty,” adopted by the General Assembly in its resolution 2222 (XXI)), opened for signature on January 27, 1967, entered into force on October 10, 1967, 98 ratifications and 27 signatures (as of January 1, 2003);

� The Agreement on the Rescue of Astronauts, the Return of Astronauts and the Return of Objects Launched into Outer Space (the “Rescue Agreement,” adopted by the General Assembly in its resolution 2345 (XXII)), opened for signature on April 22, 1968, entered into force on December 3, 1968, 88 ratifications, 25 signatures, and 1 acceptance of rights and obligations (as of January 1, 2003);

� The Convention on International Liability for Damage Caused by Space Objects (the “Liability Convention,” adopted by the General Assembly in its resolution 2777 (XXVI)), opened for signature on March 29, 1972, entered into force on September 1, 1972, 82 ratifications, 25 signatures, and 2 acceptances of rights and obligations (as of January 1, 2003);

� The Convention on Registration of Objects Launched into Outer Space (the “Reg- istration Convention,” adopted by the General Assembly in its resolution 3235 (XXIX)), opened for signature on January 14, 1975, entered into force on September 15, 1976, 44 ratifications, 4 signatures, and 2 acceptances of rights and obligations (as of January 1, 2003);

� The Agreement Governing the Activities of States on the Moon and Other Celestial Bodies (the “Moon Agreement,” adopted by the General Assembly in its resolution 34/68), opened for signature on December 18, 1979, entered into force on July 11, 1984, 10 ratifications and 5 signatures (as of January 1, 2003).

These acts constitute the bulk of what is referred to as “space law,” intended as that branch of public law that deals with activities which occur outside the earth’s atmosphere. From a practical point of view, the effect of these treaties is somewhat limited. The main reasons for their ineffectuality is that they mostly deal with issues of principle and not with the day-to-day activities of aerospace companies (d’Angelo, 1994).

The first of these acts (“Outer Space Treaty”) already includes article VII, which concerns third-party liability and states that: “Each State Party to the Treaty that launches or procures the launching of an object into outer space, including the moon and other celestial bodies, and each State Party from whose territory or facility an object is launched, is internationally liable for damage to another State Party to the Treaty or to its natural or juridical persons by such object or its component parts on the earth, in air or in outer space, including the moon and other celestial bodies.”

144 RISK MANAGEMENT AND INSURANCE REVIEW

That basic rule was even enlarged upon in the “Liability Convention,” according to which the signatory states are responsible for all acts and omissions of their government agencies and of all their natural or juridical persons. Article II of the “Liability Conven- tion” states that: “A launching State shall be absolutely liable to pay compensation for damage caused by its space object on the surface of the earth or to aircraft flight.” There is no limit to the amount of indemnity, but compensation is restricted to damage caused directly by space objects. In addition, damage on the earth is clearly distinguished from damage in outer space. The first applies if a space object inflicts damage on the surface of the earth or to aircraft in flight. In such a case the liability of a launching state shall be absolute. However, liability for damage to other space objects in outer space is based on fault (Articles III, IV, VI). In consequence such regulations of space law usually cause the necessity of buying an insurance policy against third-party liability. Also, treating dam- age on the earth and damage in outer space differently is very important when assessing the liability risk, because, according to Kowalewski (2002), the intra-space liability based on fault creates a less-intensive risk of third-party liability.

Moreover, this distinction in space law also requires a definition of where “outer space” starts. Here there are many different opinions, and this has created both sci- entific and legal problems. Simply speaking, outer space begins where airspace finishes (Antonowicz, 1998). Another definition is that outer space begins at the lowest altitude at which it is technically feasible for a satellite to orbit the earth, which is currently about 80 kilometers above sea level (Space Flight and Insurance, 1992). According to this definition, the true birth of space flight was in 1942 when a German A-4 (also called V2) rocket was launched, because its altitude exceeded 80 kilometers. Another source (Encyklopedia Geograficzna Świata, 1997) announces that space begins at about 180 kilo- meters, which is where the density of atmosphere becomes so thin that it is possible for a few days’ free flight around the earth. Although there is no clear-cut lower limit of outer space, international practice assumes that outer space “begins” at the altitude of about 100 kilometers above see level (Antonowicz, 1998).

The compensation provided for in the “Liability Convention,” depends on the identifica- tion of the space object that is responsible for the damage. It is to assure that such identifi- cation is possible that a “Registration Convention” demands that each state launching an object into outer space register the said object. If it is possible to confirm who launched the given space object, the injured party can claim its compensation on the basis of principles given in the “Liability Convention” (Articles VIII–XX).

Damages inflicted on third parties occur more often on the earth. During take-off, there is a possibility that the launch vehicle or its parts (e.g., external tanks, strap-on boosters) can cause damage to any objects on the ground, sea, or to aircraft in flight. For this reason, satellites are usually launched in a seaward direction, sometimes indeed from a platform on the sea (e.g., a Sea Launch rocket). Shipping lanes nearby and airspace in the region of the launch are closed during launching time. If a launch vehicle deviates from its nominal trajectory and threatens to cause damage, it can be blown up by a built-in self-destruction device, thus minimizing the risk of damage. The most dangerous are those accidents that arise on the launch pad or within a minute or thereabouts of take-off. This happened in 1986 when a Titan rocket exploded at a height of only 240 meters, destroying both the launch pad and the launch facilities. In another case a farmer from Georgetown in Texas had a 500-pound fuel tank from a Delta II booster rocket land nearly intact just 150 feet from his house (Coffin, 1997). Other examples include:

THE COLUMBIA SPACE SHUT TLE TRAGEDY 145

1. the failure of a Long March 3B in 1996, which pitched over before clearing the launch tower. It crashed into a hillside 22 seconds into flight, killing at least 100 people and destroying the attached Intelsat 708 satellite (Anselmo, 1999);

2. the second stage of a Thor Able Star rocket fell to the ground in Cuba and killed a cow—the U.S. Government had to pay to Cuba US$2 million in compensation, thus creating one of the more expensive cows in history (Bulloch, 1988);

3. the failure of a Proton launcher on July 7, 1999, which resulted in an 80-ton rocket fragment plummeting to the ground, 6 miles from the town of Salamalkol (Kazakhstan), with a further 440-pound piece falling into a yard of a home in a nearby village—Kazakh authorities presented a claim to the Russian Government in the amount varying between US$270,000 and US$288,000;

4. another failure of a Proton rocket on October 27, 1999, 3 minutes 40 seconds into its flight, with the reported claim paid by Russia to Kazakhstan in the region of US$400,000 (for these and more examples of accidents, see Schmid, 2000);

5. at least 21 people were killed in August 2003 in Alcantara (Brazil) after the explosion of a VLS-3 rocket on the launch pad. The rocket booster was mistakenly ignited during tests, three days prior to the scheduled launch.

It is also possible during the operation of spacecraft for harm to be inflicted on third parties. Damages in outer space are usually connected with either a collision or through electromagnetic interference in transmissions of one satellite or terrestrial radio links caused by the system of another satellite. However, there is no doubt that a guilty party is obligated to compensate for that damage.

A spacecraft could suffer damage (both partial and total loss) as a result of collision with another object. A crash is possible with three kinds of objects:

� with another operating satellite; � with space debris; � with a heavenly body such as a meteor, in which case there would be no liability.

The chance of a collision between two operating spacecrafts is small. These objects are under the constant control of ground stations that track their orbits. It has been rec- ommended for several years that satellites that have reached the end of their working life-span be moved away from their geostationary orbit. Satellites from low orbits are usually de-orbited. They partly or completely burn up in the atmosphere, with any debris theoretically falling into oceans. One example of a space object being treated in this way was the Space Station MIR, taken out of commission in 2001. Other satellites are shifted to higher orbits. In the second case the altitude increase should be at least 150 kilometers. The fuel required for that operation is equivalent to the amount needed for six weeks active station-keeping (Blassel, 1985).

Human activity in outer space has resulted in the appearance of many objects orbiting the earth. The majority no longer serve any useful purpose—old satellites, fragments of rockets—but are a danger to functioning spacecrafts. One example occurred in August 1997, when a 500-pound discarded rocket motor floating in earth’s orbit passed within 2.5 kilometers of an ozone-measuring satellite worth tens of millions of dollars. NASA

146 RISK MANAGEMENT AND INSURANCE REVIEW

alerts its space shuttles of a possible collision when any other object comes within 50 kilometers of the orbiters (Coffin, 1997).

Article II of the “Registration Convention” imposes on launch operations the obligation to catalogue all objects sent into space. Since 1957 about 9,000 objects have been logged that are still being tracked. More than 100,000 bits of debris are still in space that are too small to follow. Such debris includes pieces of aluminum chuffed from satellite boost stages, blobs of liquid metal coolant that leaks from discarded space reactors, debris resulting from satellite explosions, and lens covers and other hardware discarded during normal satellite operations. Some of this material will remain in earth orbit for hundreds or even thousands of years (Ailor, 2000). However, only 7 percent of the registered objects are still functioning—the rest are nonfunctional satellites (20 percent), rockets’ upper stages (16 percent), remains after missions (12 percent), and different fragments (45 percent). This means that over 90 percent of objects sent into outer space are now nonfunctional debris. Space (orbital) debris is technically defined as any man-made earth-orbiting object, which is nonfunctional with no reasonable expectation of assuming or resuming its intended function or any other function for which it is or can be expected to be authorized, including fragments and parts thereof (Flury, 1999).

Currently, the possibility of an operational satellite being damaged or destroyed by space debris is small (estimated by actuaries at about 0.01 percent), but as the amount of debris in space increases, the possibility of an operational satellite being hit is rising. This process is irreversible, since the cleaning-up of space is economically (and also technically) unfeasible. Most space debris is located in orbital regions that are frequently used for a multitude of applications (low orbits: 800 to 1,600 kilometers and geostationary orbit of about 36,000 kilometers above the earth’s surface).

For large close-to-earth orbiting spacecraft and for space debris there is a risk of a fall to earth. The lower the orbit and the greater the mass, the greater the chance of a reentry. A satellite falling to the earth has the same effect as a natural meteor. When it passes through the atmosphere, huge heat and pressure develops and the object is broken up into numerous pieces, most of which are completely burnt up. Only a very few large pieces survive to reach the ground. Some examples of reentries from outer space:

1. the spent stage of a Saturn V rocket, weighing about 22 tons, which fell into the Atlantic Ocean east of the Azores in January 1978;

2. the American Skylab, weighing approximately 80 tons, crashed over the western coast of Australia in July 1979 (Space Flight and Insurance, 1992).

However, in reality, despite the large size of these objects, the risk of damage to the earth is quite low—over two-thirds of the earth’s surface is sea and much of the land is sparsely populated.

What causes more concern is the environmental damage that can be caused by space- craft with nuclear power generators on board. On January 24, 1978 the Russian satellite Cosmos 954 crashed in Northwest Canada, contaminating large areas with radioactivity. Based on the provisions of the “Liability Convention” and general principles of inter- national law, a claim in the total amount Can$6.04 million was submitted, although the matter was settled some time later following negotiation, in the amount of Can$3 million. There are still spacecraft that use nuclear materials for power supplies. This constitutes a serious risk.

THE COLUMBIA SPACE SHUT TLE TRAGEDY 147

The service and/or repair of spacecrafts in orbit could cause liability of the owner of the device for potential damage. It is unclear what would happen if, during replacement of a broken part, the astronaut-mechanic destroyed the repaired module. How can companies that have spent huge sums of money in the manufacturing of such equipment protect themselves against the risk of sharing multipurpose platforms or space stations? How can the “earth” (national) law be applied to these situations? International space law has not solved this problem yet. This matter should engage not only lawyers, but also other interested parties, including the insurance community.

SPACE THIRD-PARTY LIABILITY INSURANCE IN THE WORLD INSURANCE MARKET The need to procure third-party liability insurance is based on protection against fi- nancial claims resulting from certain fundamental principles of international space law (mainly the “Outer Space Treaty” and the “Liability Convention”) as well as national leg- islation, executive orders, administrative regulations, and judicial decisions that control or otherwise influence the conduct of activities in space (Meredith, 1992). The require- ment for and scope of liability cover is dependent on the Launch Services Contract with the launching agency. In some cases the satellite owner is responsible for the purchase of insurance, but the majority of launch suppliers now include the arrangement of the appropriate coverage as part of the launch services supplied by them.

In general, liability insurance covers the insured against potential claims and ensures compensation for the victim. Therefore, liability insurances fulfill a double protection function. Space third-party liability insurance has the same purpose.

It covers the legal liability arising from damage to a third party during the preparations for launch, the lift-off itself, in-orbit operations of a satellite program, and finally the reentry. This type of insurance will provide compensation in the event of personal injury and property damage to third parties, both on the ground and in space, caused by the launch vehicle sections or the satellite. So the space third-party liability insurance applies to damages to a third party in connection with such events as: falling of a satellite or a rocket or elements thereof on the ground, fire during ignition, explosion of a satellite in orbit, collision with another spacecraft, etc. (Zocher II, 1988; Zocher IV, 1988). The launch pad is usually not covered. Neither is damage to payloads, since there is often a clause in the underlying contracts in which all parties agree to a cross-waiver of liability. According to Pino (1997) this applies also even in the case of gross negligence. Therefore, insurance covers the period from the delivery of a spacecraft to a launch pad till the day of expiration of that policy or the destruction of the satellite, whichever comes first. Contracts are extended to the end of a spacecraft’s life.

The launch service providers typically purchase third-party liability insurance for the launch of a satellite and for a set period thereafter. They will add the satellite operator to the liability insurance they hold as an additional named insured. The satellite operator will also occasionally purchase in-orbit third-party cover, which comes into operation when the launch coverage expires. This insurance is taken out either to comply with leg- islation in certain countries, or for the satellite operator’s own peace of mind. Sometimes producers, launching states, or other related organizations could be coinsured.

Exclusions that are typically applied to a third-party liability policy, include (Margo, 2000):

148 RISK MANAGEMENT AND INSURANCE REVIEW

� war risks; � claims caused by radioactive contamination of any nature whatsoever; � noise, pollution, and related risks; � any obligation of the insured to his employees or any obligation for which the

insured or any carrier as his insurer may be liable to his own employees, under any workers’ compensation, death, or disability benefits law, equal opportunity laws, or under any similar law;

� any damages to the property of the insured; � claims resulting from an interruption in telecommunications service to satellites,

whatever cause thereof; � liability of any insured as a manufacturer; � claims made for the failure of the spacecraft to provide communications service.

The limits recently purchased vary from around US$60 million to US$500 million. For example, in the United States, the government has renewed legislation that limits com- mercial operations liability for damage caused by a launch failure to US$200 million, with the U.S. government responsible for the balance of up to US$1.5 billion in liability specified by international treaties (Pagnanelli, 2001).

Rates differ considerably. They are affected by trends in the overall liability market and the capacity required as well as specific liability issues. In the context of the launch (14 percent to 18 percent of the sum insured) and in-orbit (2 percent to 4.5 percent of the sum insured) premiums, liability premiums are relatively small amounts and are typically at a level of around 0.1 percent (per year) of the required limit of liability (Space Insurance Briefing, 2001). However, when Russians protected themselves against the failure of the falling of the MIR Station into the Pacific ocean (March 23, 2001), they had to pay about US$1 million premium for US$200 million limit of responsibility. The high level of premium required could have shown the degree of confidence of the insurance market in the reliability of MIR.

CONCLUSIONS Thus far there have been only a few cases of third-party liability for space losses. It should also be noted that there has never been a substantial claim on a space liability insurance policy. It remains to be seen if this type of coverage would remain available if a major accident was to occur. The tragedy of the Columbia space shuttle shows that potential damage could be enormous (if the catastrophe had occurred above a city). The debris of the orbiter fell on a sparsely populated area near the Texas/Arizona border. In total, NASA received 66 claims for property damage and loss of cattle, totaling US$500,000. The corridor of debris passed 15 miles south of Houston and Fort Worth. However, it also has to be said that the debris of the space shuttle Columbia did not hit or hurt a single person. According to Mr. Pastorek, NASA self-insures what it flies (Stahler, 2003).

So again it should be emphasized—with the development of space transportation—both commercial and noncommercial (governmental, scientific, etc.)—issues of risk manage- ment are very important in view of the considerable financial commitments of launch

THE COLUMBIA SPACE SHUT TLE TRAGEDY 149

participants and the enormity of damages that may occur. In addition to the risk involved in the loss or failure of spacecraft that we have frequently observed, space activities cre- ate exposure to potentially “astronomical” (or even “out of this world”) liability to third parties injured by the malfunctioning spaceship or rocket boosters.

REFERENCES Ailor, W., 2000, New Hazards for a New Age, Crosslink, 1(1): 20-23. Anselmo, J., 1999, Cox: Companies Broke Law—and Knew It. Aviation Week & Space

Technology, 150(22): 30-31. Antonowicz, L., 1998, Podręcznik prawa międzynarodowego (Warsaw: Wyd. Prawnicze

PWN). Blassel, P., 1985. Space Projects and the Coverage of Associated Risks. The Geneva Papers

on Risk and Insurance, 10(35): 51-86. Bulloch, C., 1988, Commercial Space Launches. Liability Questions Resolved at Last.

Space Markets, Winter: 211-14. Coffin, B., 1997, Lost in Space. Best’s Review/Property-Casualty Insurance Edition, 98(7):

68-72. d’Angelo, G., 1994, Aerospace Business Law (Westport: Quorum Books). Daouphars, P., 1992, L’assurance des Risques Spatiales’, in: Kahn, P., L’exploitation Com-

merciale de l’Espace (Paris: LITEC). Jelonek, A., ed., 1997, Encyklopedia Geograficzna Świata (Krakow: Tom VIII—Wszechświat,

OPRES). Flury, W., 1999, Space Debris a Hazard to Operational Spacecraft? In: Commercial and

Industrial Activities in Space—Insurance Implications (Trieste: Generali), pp. 41-49. Kowalewski, E., 2002, Istota ubezpieczenia odpowiedzialności cywilnej, Prawo Asekura-

cyjne, 3: 3-13. Margo, R., 2000, Aviation Insurance. The Law and Practice of Aviation Insurance, Including

Hovercraft and Spacecraft Insurance, 3rd edition (London, Edinburgh, Dublin: Butter- worths).

Meredith, P., and G. Robinson, 1992, Space Law: A Case Study for the Practitioner: Imple- menting a Telecommunications Satellite Business Concept (Amsterdam: Martinus Nijhoff Publishers).

Pagnanelli, B., 2001, Space Insurance Towards the Next Decade. In: Commercial and In- dustrial Activities in Space—Insurance Implications (Trieste: Generali), pp. 25-33.

Pino, R., 1997, With the Continued Development of Space, the Satellite Industry will En- counter new Frontiers in the Legal Claims Area. In: Commercial and Industrial Activities in Space Insurance Implications (Trieste: Generali), pp. 189-97.

Schmid, T., and D. B. Downie, 2000, Assessing Third Party Liability Claims, In: The 9th International Space Conference (London: IBC).

Space Flight and Insurance, 1993, 2nd edition (Munich Re).

Space Insurance Briefing, 2001, (London: Marsh Space Projects Ltd.). Stahler, W., 2003, Of New Risks, Unknown Risks and Uncertainty. Risk Management, 33:

1-4.

150 RISK MANAGEMENT AND INSURANCE REVIEW

Zocher, H., 1988, Neuere Internationale Entwicklungen in der Raumfahrt und ihrer Versicherung (II), Versicherungswirtschaft, 43(2): 147-55.

Zocher, H., 1988, Neuere Internationale Entwicklungen in der Raumfahrt und ihrer Versicherung (IV), Versicherungswirtschaft, 43(4): 284-90.