Minogue, Katie; Wright, Tamara --- "What would happen if KITT, the Batmobile and Herbie were on our roads today?: Australia s transport accident schemes and driverless cars" [2019] PrecedentAULA 34; (2019) 152 Precedent 45

---

WHAT WOULD HAPPEN IF KITT, THE BATMOBILE AND HERBIE WERE ON OUR ROADS TODAY? AUSTRALIA’S TRANSPORT ACCIDENT SCHEMES AND DRIVERLESS CARS

By Katie Minogue and Tamara Wright

While we might be some time away from riding along in Knight Rider's Kitt while he throws dry, pithy one-liners at us, the introduction of fully autonomous vehicles to our roads is not as far away as we may think.

It comes as no shock that autonomous vehicle technology in the form of adaptive cruise control, auto lane-change, emergency braking and self-parking are utilised in cars on our roads today. However, it is less commonly understood that fully driverless technology is already being tested on Australian roads. Vehicle manufacturers as well as ride-share companies are investing heavily in the roll-out of fully automated vehicles. Recently Bosch became the first manufacturer granted a permit to test automated vehicles on high-speed rural Victorian roads, with many others set to follow suit.^[1]

With over 90 per cent of motor vehicle accidents caused by human error, there is little argument that autonomous vehicle technology will substantially reduce death and injury on our roads.^[2] However, removing human drivers will not eliminate accidents entirely. Accidents may be caused by autonomous vehicle design faults, software errors or by interactions between driverless cars and human-driven cars, pedestrians, cyclists or motorcyclists. As the law stands today, an injured person may be treated differently under Australia’s motor accident injury insurance (MAII) schemes depending on whether they are injured by a human driver or a vehicle operating in autonomous mode. MAII schemes were not designed to contemplate an autonomous system ‘driving’ a motor vehicle and significant reform is needed to ensure fairness and transparency for all road users.

In its Motor Accident Injury Insurance and Automated Vehicles discussion paper released in October 2018, the National Transport Commission (NTC), the statutory body tasked with making transport reform recommendations to government, recognised that reform is needed. The purpose of the discussion paper is to identify barriers to accessing compensation under current MAII schemes for people injured in an accident involving an automated vehicle. It also seeks feedback on whether existing MAII schemes, or alternative insurance models, should provide cover in these circumstances. Twenty-two parties have made submissions including state law institutes and councils, law firms, insurance councils and insurance companies.^[3] Feedback received from the discussion paper will be used to develop recommendations for consideration at the Transport and Infrastructure Council (the Council) meeting in May 2019.^[4]

MAII SCHEMES IN AUSTRALIA

Each Australian state and territory currently operates a MAII scheme which requires every registered vehicle to have compulsory third-party insurance for injuries and death resulting from transport accidents. Vehicle owners fund the schemes by insurance premiums paid through annual registration charges.

In Victoria, New South Wales and Tasmania, injured parties are able to access no-fault entitlements in addition to common law damages (hybrid schemes). In Western Australia, South Australia, the Australian Capital Territory and Queensland, only common law damages are accessible, which relies on an injured person establishing negligence against an insured party (fault-based schemes).^[5] Only the Northern Territory scheme is a completely no-fault scheme. The National Injury Insurance Scheme (NIIS) provides additional entitlements for catastrophically injured parties who have not received common law compensation.

THE DISCUSSION PAPER

The NTC has identified three key issues that arise when existing MAII schemes are applied to accidents involving autonomous vehicles.

Firstly, the definitions which set out inclusion in MAII schemes are designed to apply to cars driven by humans and do not contemplate an accident involving an autonomous driving system (ADS). An injured person may not be covered by a MAII scheme if they are not injured by ‘the driving of’ a vehicle. To satisfy the definition of ‘driving’, most states require a human driver to be in control of the vehicle.^[6] This means that a person injured by an ADS may not have the same, or any access to compensation compared with someone injured by a human driver.^[7] If the injured person fell outside the MAII scheme, they would need to rely on the existing frameworks of common law negligence, an action for breach of contract or the Australian Consumer Law in order to obtain compensation for their injuries.

Secondly, the requirement to establish negligence under common law schemes becomes more difficult when the insured party is not a human owner or driver but is an ADS. An ADS is a machine and cannot be negligent.^[8] This means that complex and costly litigation may be necessary in order to identify the relevant party against whom negligence could be established. Accordingly, legislative changes are needed to ensure that there is a party clearly capable of being sued in cases where negligence has led to an ADS crash.

Finally, MAII schemes are set up so that governments, insurers and vehicle owners subsidise the system. They are not set up to ensure that the cost of ADS crashes is borne by those who control the risks (for example, manufactures and software designers).^[9] In the long term, significant MAII scheme redesign may be needed to achieve this outcome.

The NTC has approached all options for reform, guided by the overarching principle that:

‘No person should be worse off, financially or procedurally, if they are injured by a vehicle whose automated driving system was engaged, than if they were injured by a vehicle controlled by a human driver.’^[10]

The discussion paper offers a number of options for reform. These are:

• Option 1: Rely on the existing legal framework;

• Option 2: Exclude injuries caused by an ADS from MAII schemes;

• Option 3: Expand MAII schemes to cover injuries caused by an ADS;

• Option 4: Implement a purpose-built automated vehicle scheme;

• Option 5: Implement minimum national benchmarks for the scope and coverage of injuries resulting from automated vehicle accidents; and

• Option 6: Modify MAII schemes to allow single private insurers for property damage and personal injury.^[11]

Options 1 and 2

Options 1 and 2 were dismissed by most parties who made submissions, as well as by the NTC, as unlikely to meet community expectations of equality of access to compensation. They also fail the test of satisfying the NTC’s overarching principle.

Relying on the existing legal framework (Option 1) would create uncertainty for all parties as it would be unclear whether an injured person would fall under a MAII scheme or would need to rely on common law negligence, the Australian Consumer Law or contract law.

Excluding injuries caused by an ADS from MAII schemes (Option 2) would ensure that a person injured by an ADS would only be able to rely on common law negligence, the Australian Consumer Law or contract law.

Both scenarios would clearly create an unfair outcome. Consider a case where a pedestrian is hit and injured by a car that fails to stop at a pedestrian crossing. The pedestrian may be subject to completely different insurance coverage and outcomes depending on whether the car that hit her was utilising auto-emergency braking or whether the human driver applied their foot to the brake pedal. Further, whether the human driver is ‘in control’ of the vehicle at the time the brakes were applied would need to be established in order to determine whether she is included or excluded from her state’s MAII scheme. Not only does this create an arbitrary and unfair outcome, but it would involve costly and burdensome analysis of the car’s ‘black box’ data to determine whether a human being or ADS was ‘in control’ at the time of the accident.

It is worth noting that the Insurance Commission of Western Australia preferred Option 2 (excluding injuries caused by an ADS from MAII schemes) on the basis that ‘MAII schemes were developed to cover the specific risk of human driver negligence causing injury...not...vehicle manufacturer negligence and product liability risks.’^[12] This is not supported by other parties’ submissions. Most parties accept that motor vehicles are unique due to their potential to cause great harm, and are not easily likened to, for example, a household appliance. This agreed understanding underpins the consensus that personal injury insurance for motor vehicle accidents should be compulsory.^[13]

Options 4 and 6

A purpose-built automated vehicle scheme (Option 4) and modifying MAII schemes to allow single private insurers for property damage and personal injury (Option 6) were also dismissed by most parties who made submissions. These options would require a comprehensive overhaul of existing MAII schemes which would be both costly and premature at this early stage. However, many parties submitted that these options may become more viable once autonomous vehicle usage has increased and safety risks from early deployments could be taken into account.

Options 3 and 5

The majority of stakeholders have agreed that the most appropriate option at this stage is Option 3: to expand MAII schemes to cover injuries caused by an automated driving system. Under this option, MAII laws would be amended to remove barriers to accidents/injuries caused by ADSs. Injured people would have access to compensation and benefits regardless of whether the injury was caused by an automated vehicle whose ADS was engaged or by a vehicle controlled by a human driver.^[14]

Several parties also submitted that in addition to Option 3, minimum benchmarks (Option 5) may be helpful in guiding the insurance model so as to ensure consistency across states. Possible minimum benchmarks could include:

1. Injured people are required to approach only one scheme and/or one insurer;

2. The minimum level of coverage for automated vehicles is third party personal injury insurance;

3. Sufficient recovery mechanisms are implemented to ensure that schemes and/or insurers can recover from those in the automated vehicle supply, operation and maintenance chain for product defects and failures; and

4. It is compulsory for automated vehicles to obtain CTP insurance.^[15]

It is clear that Option 3 has the most benefits, especially in the short to medium term as autonomous technology is gradually rolled out. Benefits of extending MAII scheme coverage to all injured road users include consistency and parity of coverage, enabling the injured person to deal with the usual insurer in their jurisdiction and ensuring that the responsible entity is experienced in claims management and is subject to MAII regulatory oversight.^[16]

The Law Institute of Victoria (LIV) emphasises in its submission that a single point of access for individuals is crucial.^[17] Being excluded from MAII schemes may subject an injured person to additional legal costs and delays. Also, different damages caps apply under the Australian Consumer Law and under MAII schemes. Option 3 would eliminate these inequities and fulfil the NTC’s overarching principle that no person should be worse off procedurally or financially if an injury is caused by ADS as opposed to a human driver.

Several legislative changes would need to occur in order to implement Option 3. These include:

1. broadening the definition of ‘driving’ beyond human driving to include ADS operation;

2. extending indemnification beyond owners and drivers of registered vehicles to include an ADS when engaged; and

3. ensuring that there is an insured party (for example, the manufacturer) that can be claimed against in negligence.^[18]

While Option 3 presents the most streamlined, efficient and fair solution in the short to medium term, it is not without its challenges.

Cost shifting

MAII insurance premiums are paid by vehicle operators, and the cost of the scheme is also borne by insurers, government and taxpayers. If this model does not change with the advent of autonomous vehicles, the cost of personal injury caused by a failure of an ADS will not be borne by the negligent party responsible for the defect.^[19] Potentially negligent parties include manufacturers, software providers, repairers, installers, and infrastructure and telecommunications providers.

This gap should be addressed by ensuring that insurers have the right to recover the costs of ADS crashes from at-fault third parties. The LIV has submitted that ‘when properly exercised, recovery rights provide sufficient protections for operators and insurers, particularly given we should be seeing a reduction in road accidents in the longer term as a result of the technology due to less claims’.^[20]

On the other hand, it is argued that requiring insurers to establish fault against a specific party in the long supply chain (from manufacturers to repairers) would be burdensome and complex. It may result in protracted and costly disputes between insurers and parties. This runs the risk that the scheme becomes unworkable and that costs would not be recovered, resulting in significantly higher premiums for all vehicles.^[21]

Reinsurance pool

The NTC has therefore proposed a modification of Option 3 that would address these cost-shifting issues: to create a national reinsurance pool.^[22] A reinsurance pool would involve contribution by all parties in the supply chain to a pool of funds through risk-rated premiums. The MAII insurer would then have a right to recovery from the reinsurance pool.^[23]

The major problem with this proposal, identified by many parties making submissions, is that when so little is known about risk at this early stage of autonomous vehicle deployment, setting risk-rated premiums would be very difficult to achieve. If the risk-rating system is not correctly weighted, wrongdoers would not be encouraged to create change. Additionally, the ‘pool’ system may require faultless parties, who have never contributed to any defect causing an accident, to continually subsidise parties who have contributed to a defect. Further, many parties agree that the reinsurance pool is premature, unwieldy and administratively burdensome. The Insurance Council of Australia posits that a ‘more prudent approach would be for schemes to respond in accordance with real experience and information gathered. This will facilitate incremental change, subject to the experiences of each state, and allow schemes to respond appropriately to needs as they arise.’^[24]

Right of recovery

As an alternative to the reinsurance pool model, most parties suggest that insurers can utilise existing recovery rights against the autonomous driving system entity (ADSE) or any ‘at fault’ third parties. The ADSE is defined as the party responsible for certification before a car is released to market, as well as the entity responsible for compliance with safety and traffic obligations.^[25] Several parties submitted that the ADSE is the entity best placed to take action against risk of defect and therefore should be responsible for the cost of damages.^[26]

Several parties took this idea a step further and proposed a strict legislated right of recovery against the ADSE in situations where the ADS was engaged at the time of the accident and was found to be wholly or partially at fault, or where there is a hardware or software failure. They argued that a legislated right creates a clear and certain line of recovery. As QBE submitted:

‘If properly designed, this will support the overarching principle and assist to avoid protracted disputes between insurers, ADSEs, and component manufacturers that would otherwise add unworkable delays and friction costs to the scheme...A direct recovery pathway would, along with the safety assurance system, also encourage ADSEs to prioritise holistic vehicle safety through proper design, testing and after-sales care. ADSEs would be free to pursue other entities in the vehicle supply chain as they do today.’^[27]

This approach is supported by the outcome of the NTC’s Safety Assurance for Automated Driving Systems: Consultation Regulation Impact Statement, released in May 2018, which stipulates that an ADSE must have a corporate presence in Australia that can be criminally and civilly liable under Australian law and that the ADSE must also hold an appropriate level of insurance to cover personal injury, death and property damage caused by the ADS when it is properly engaged.^[28]

To provide more certainty for insurers regarding their prospects of successfully recovering costs, another suggestion is that where an accident was caused by the ADS, there should be a rebuttable presumption of fault against the ADSE.^[29] This has the advantage of sending a clear signal to the ADSE around responsibility and costs, and would relieve the injured person of the burden of proving that the injuries were caused by the manufacturer’s or ADSE’s negligence.^[30]

Based on all the submissions of stakeholders, it is likely the NTC will find Option 3, with the possibility of Option 5 minimum benchmarks in addition, to be the preferable way forward in the immediate term. The weight of the submissions support the insurer’s right of recovery as the most straightforward way to ensure that the cost of accidents is borne by the responsible party in the supply chain, rather than the impractical reinsurance pool suggestion.

ACCESS TO DATA

The discussion paper also asked whether existing processes are sufficient to enable access of autonomous vehicle data for the purposes of establishing liability in relation to a personal injury claim.

Most parties agree that reform is needed to establish a streamlined and cost-efficient mechanism for all parties to an accident (including the injured passenger) to access event data and ensure that no person is worse off financially or procedurally, in accordance with the NTC’s overarching principle. Early access should result in a reduction of complicated litigation, particularly if data is made available soon after the accident to settle questions of fault at an earlier point in time.

The owner of the data is most likely to be the manufacturer or the ADSE. As the potential defendant in any potential negligence claim, this party will have an interest in protecting the data. Allowing them to control access to the data creates a significant disadvantage to the injured person and affords the manufacturer or ADSE an inequitably powerful position. Balance must be exercised between allowing industry to collect and manage and preserve data on the one hand, with governmental oversight and access to that data as part of civil proceedings on the other.^[31]

CONCLUSION

MAII schemes provide coverage for anyone who has the misfortune to be injured on the road as a driver, passenger, cyclist or pedestrian. It is rare to find ourselves in a period where the next few months may so significantly change such a large and wide-reaching body of entitlements.

We know that autonomous vehicle technology is likely to reduce the incidence of death and injury on the road. However, the public is unlikely to accept driverless technology if they are subject to an opaque and unfair system that applies double standards depending on whether human or ‘automated system’ agencies are behind the wheel. It is critical that the NTC’s regulatory reform paves the way for equitable coverage for all under MAII schemes, resulting in greater uptake of autonomous technology and greater safety outcomes for all on the road. That would certainly be a sci-fi fantasy worth getting excited about.

Katie Minogue is a Senior Associate practicing in the area of road and work injuries at Maurice Blackburn Lawyers in Frankston. She is an Accredited Specialist in Personal Injury Law. EMAIL [e-mail].

Tamara Wright is a lawyer practising in road injuries at Maurice Blackburn Lawyers in Dandenong. EMAIL [e-mail].

Katie and Tamara have prepared numerous submissions concerning the regulation of automated vehicle technology to bodies including the National Transport Commission, VicRoads and the Federal Government. They have also appeared in front of the House of Representatives Standing Committee on Industry, Innovation Science and Resources in its inquiry into driverless vehicles.

---

^[1] Minister for Transport Infrastructure, First Automated Vehicle to Hit the Road, 21 January 2019, <https://www.premier.vic.gov.au/first-automated-vehicle-to-hit-the-road/>.

^[2] B Walker Smith, ‘Human Error as a Cause of Vehicle Crashes’, The Standford University Center for Internet and Society (18 December 2013) <http://cyberlaw.stanford.edu/blog/2013/12/human-error-cause-vehicle-crashes> .

^[3] National Transport Commission, Submissions for NTC Discussion Paper – Motor accident injury insurance and automated vehicles, <https://www.ntc.gov.au/submissions/history/?rid=167380&pid=11983>.

^[4] National Transport Commission, Motor Accident Injury Insurance and Automated Vehicles (October 2018) 1.

^[5] Legislation is currently before the ACT Legislative Assembly which dramatically changes the CTP scheme in the ACT to a no-fault scheme. By the time this article is published the legislation is likely to have been passed.

^[6] National Transport Commission, above note 4, 1.

^[7] Ibid.

^[8] Ibid, 2.

^[9] Ibid.

^[10] Ibid.

^[11] Ibid, 3.

^[12] Insurance Commission of Western Australia, Motor Accident Injury Insurance and Automated Vehicles Submissions (19 December 2018) 2.

^[13] Law Council of Australia, Motor Accident Injury Insurance and Automated Vehicles Submissions (12 December 2018) 5.

^[14] National Transport Commission, above note 4, 43.

^[15] Motor Accident Insurance Commission (Queensland), The Motor Accident Insurance Commission response to the National Transport Commission’s Motor Accident Injury Insurance and Automated Vehicles Discussion Paper (December 2018) 9.

^[16] Department of Treasury and Finance (South Australia), Submission to National Transport Commission Discussion Paper: Motor Accident Injury Insurance and Automated Vehicles (December 2018) 3.

^[17] Law Institute of Victoria, Submission to the National Transport Commission Discussion Paper: Motor Accident Injury Insurance and Automated Vehicles (13 December 2018) 2.

^[18] National Transport Commission, above note 4, 79.

^[19] Ibid, 49.

^[20] Law Institute of Victoria, above note 17.

^[21] National Transport Commission, above note 4, 50.

^[22] Ibid, 57.

^[23] Ibid, 50.

^[24] Insurance Council of Australia, Discussion Paper: Motor Accident Injury Insurance and Automated Vehicles (12 December 2018) 5.

^[25] National Transport Commission, Changing Driving Laws to Support Automated Vehicles: Policy Paper (May 2018).

^[26] Ibid, 1.

^[27] QBE, Submission on Motor Accident Injury Insurance and Automated Vehicles (December 2018) 4.

^[28] National Transport Commission, Safety Assurance for Automated Driving Systems Consultation Regulation Impact Statement (May 2018) 88-89.

^[29] Maurice Blackburn Lawyers, Submission in Response to the Motor Accident Injury Insurance and Automated Vehicles Discussion Paper (December 2018) 8.

^[30] National Transport Commission, above note 4, 80.

^[31] Law Institute of Victoria, above note 17, 4.