Drones: Emerging commercial potential, emerging liabilities

  • Print
  • Connect
  • Email
  • Facebook
  • Twitter
  • LinkedIn
  • Google+
By Carl X. Ashenbrenner | 15 June 2015

If life imitates art more than art imitates life, then be prepared to duck because the drones are coming. In 2014, more than 10 major motion pictures featured drones and even more movies were filmed using drones. There is currently a Broadway play about a drone pilot (Grounded with Anne Hathaway), a high-concept music LP (called Drones by Muse), and even the first graffiti made by a street artist using a drone. Drones have clearly taken center stage and the property/casualty (P/C) insurance industry is responding.

Rise of the drones

Drones, also referred to as remotely piloted aircraft (RPAs), unmanned aerial vehicles (UAVs), or unmanned aircraft systems (UASs) when discussions include payload and the land-based control system, come in a variety of shapes and sizes. While “unmanned aircraft,” such as model airplanes, have been in existence virtually since the beginning of commercial air travel, the use of drones has grown exponentially in recent years, which is due both to publicity from their effective use by the U.S. military and to their improved technology.

Drone usage by the U.S. military began during the Cold War and continued through the Vietnam War and the Balkans crisis in the 1990s, primarily for reconnaissance missions. The employment of drones and the attention paid to them by the American public changed dramatically subsequent to the terrorist attacks of September 11, 2001. After the attacks, President George Bush signed a directive to authorize weaponized drones, some known as “Predators,” for use in eliminating suspected terrorists. The U.S. military and intelligence agencies continue their high-profile (but controversial) use of drones today, particularly in Pakistan and Afghanistan.

In regard to technology, drones in recent years have become cheaper, lighter, more technologically advanced, and increasingly easy to use (many are controlled through iPads). Built-in cameras can take 12 megapixel stills and video at the 1080p high-definition standard. Newer models come with built-in Global Positioning System (GPS) technology and motorized mounts that stabilize cameras while letting them rotate in several directions. For all the sophisticated equipment on board, drones are relatively inexpensive—a new Phantom 3 made by Da-Jiang Innovations (DJI) is selling for less than $1,000.

Uses for drones in the future will range among a multitude of industries, including distribution, film, energy, engineering, pleasure, and agriculture. For example, current forecasts estimate agriculture will employ the majority of drones in the near future. Agriculture drones are expected to replace tasks performed by walking through fields as well as livestock surveillance and crop-dusting. Precision agriculture relates to a maximization of yields combined with a reduction in input costs (fertilizer, herbicide, tilling, etc.). Drones are expected to quickly scan fields and provide recommendations regarding inputs. In the future, drones may also be able to provide many of the necessary inputs in the fields in the correct places with the recommended amounts. Collectively, this is expected to lower input costs as well as produce lower nitrate runoff, which can contaminate waterways.

Current rules and regulations

In the United States, the Federal Aviation Administration (FAA) is charged with regulating UASs. Since 1990, the agency has allowed limited use of drones for critical public missions such as firefighting, disaster relief, border patrol, and law enforcement. For small UASs, defined as those weighing 55 pounds or less, there are two sets of rules—one for personal use (for hobbyists) and one for commercial use (in furtherance of a business). Basic rules for hobbyists include keeping the drone (1) within visual line of sight of the operator, (2) less than 400 feet above ground level, and (3) five miles from an airport. In regard to commercial usage, the FAA released a proposed set of regulations for small UASs in February 2015. Some key limitations include the following:

  • No operations beyond visual line of sight (VLOS)
  • No operations over any person not directly involved in flight operation
  • Operators are required to obtain an unmanned aircraft operator certificate, be at least 17 years old, and be vetted by the Transportation Security Administration
  • Operations are limited to daytime hours only (sunrise to sunset)
  • Operations are limited to 500 feet above ground level (AGL)

The proposed regulations are anticipated to become law in 2017 after a brief exposure and discussion period. Until the final rules are approved, operators must receive approval from the FAA to operate a drone for commercial purposes. Approval comes in the form of a “Section 333” exemption under the FAA Modernization and Reform Act of 2012. According to the FAA's website, as of May 2015, over 450 exemptions have been granted. The majority of these exemptions relate to real estate imaging, inspections of infrastructure, surveying, agriculture, and filmmaking. In most cases, these exemptions were granted because the use of drones improved the safety of the work involved.

In its 2011 Aerospace Forecast, the FAA estimated 30,000 small drones would be in use for commercial purposes by 2030. In its 2014 Aerospace Forecast, the FAA offered a shorter term projection of 7,500 drones in commercial use by 2018. The Association of Unmanned Vehicle Systems International (AUVSI) in its report “The Economic Impact of Unmanned Aircraft Systems Integration in the United States” bullishly anticipates annual sales of 160,000 drones for commercial use (primarily agriculture) by 2025. The growth rate of drone usage may be even faster than these projections as Amazon and Google, two of the largest potential users of drones for package delivery, have recently commented on a potential shift in the FAA’s regulatory stance. They expect the FAA to announce an initiative to study drone flights beyond the line of sight of operators. This one change could materially increase the use of drones in the near-term, which is due to the many opportunities that would become available.

What is the impact on the P/C insurance industry?

Besides providing insurance for the production and use of these drones, insurance companies are also integrating drones into their own operations, transforming how they do business. For example, drones can be used to underwrite risks by carrying out surveys, inspections, and risk assessments for buildings and properties where access for human beings would be difficult, unsafe, or expensive. In regard to claim management, drones can be used to improve damage assessment—particularly in times of catastrophic losses.

Section 333 exemptions have already been granted to American International Group (AIG), Erie Insurance Group, State Farm, The Travelers Companies, and the United Services Automobile Association (USAA). These exemptions allow the insurers to use drones specifically for “aerial data collection” and also allow the companies to conduct research and development to explore new ways to use drones to support company needs. Crop insurers are also beginning to use drones as a loss adjustment and pre-acceptance inspection tool. Allstate Insurance Company is also part of a consortium dedicated to the research of drone usage.

Coverage for aviation risks is typically excluded in most general liability policies. This coverage is also typically excluded in reinsurance contracts not specifically covering aviation risks. This is largely due to the high-severity risk associated with aviation risks as well as to the difficulty in pricing for the exposure of industries not focused on aviation but that have a small amount of aviation risk.

Because of the increasing usage of drones, there is a growing segment of liability that is not currently covered in traditional general liability policies. Some property/casualty insurers are addressing coverage by extensions to general liability policies. Many specialty aviation insurers, including Global Aerospace, have produced drone-specific policies to cover a broad range of exposures, including physical damage. We discuss below some of the risks associated with drones and possible rating methodologies.

Forecasting loss costs is difficult with any new product. Currently, there is a lack of historical exposure information and little claim data for drones. In order to estimate the loss costs, one would first need to consider the risks associated with drones. If we exclude malicious acts, there are many risks associated with drones:

  • Property damage to the drone itself
  • Property damage to drone accessories (cameras, applicators, other payload)
  • Theft of drone/accessories
  • Liability due to property damage caused by drone (accidentally flying through window, etc.)
  • Liability due to bodily injury caused by drone (flying drone into person or object that causes bodily injury)
  • Libel or slander due to privacy issues from data collection
  • Cyber liability due to theft of data collected by drone
  • Product liability from the drone manufacturer

The magnitude or severity of the liability associated with drones can range from a small amount to an extremely large amount. While many of the worst-case scenarios may be caused by malicious acts, it would be possible for some to occur because of accidental or careless acts of the drone operator. One of the worst-case scenarios would be a drone interfering with the takeoff or landing of a large commercial aircraft. While regulations are being written to avoid this terrible scenario, it is not out of the range of possible events. Other scenarios could include causing large traffic accidents, etc.

Once an incident has happened, the assignment of liability to responsible parties would occur. The aviation insurance market is well versed in assignment of liability among operators, manufacturers, and owners/lenders. However, new laws, regulations, and legal theories may complicate the process. In most aviation events, the airline is held to strict liability and needs to find contributing parties (manufacturers, airports, traffic control, other aviation parties) during the settlement of the claim. This may be similar for drones although there may be a systemic risk associated with manufacturing of drones, production of software to operate drones, leasing to irresponsible parties, etc. This could be further aggravated by many operators not wishing or being unable to purchase insurance in the first place.

The rating categories for drones may be classified under various terms such as:

  • Model of drone (size, cost).
  • Usage: Commercial versus personal. Commercial may be split by industry, such as:
    • Agriculture
    • Energy
    • Insurance adjusting
    • Surveying
    • Film
  • Location/territory: Greater concentrations of property and people should have a higher risk.
  • Experience of user: Users of drones are currently required to obtain an FAA permit, but there may be alternative licenses in the future. It is recommended that insurers capture information about users for use in predictive models to assist in the rating process.
  • Owner/licensee/lease: We expect a wide variety in the usage and ownership of drones. For example, an implement dealer may lease drones to local farmers. The operator may be provided by the dealer, but may also be a farmer with the necessary FAA license. Independent firms may also provide services. If a company hires a drone operator for a task and there is a resulting event that causes a loss, who is liable? What if the contractor does not have insurance or is bankrupt?
  • Flight hours during the exposure year.

Capturing this data may not initially pay off, which is due to the limited number of claims. However, building a data set that can measure risk characteristics will be helpful in the rating process and may provide competitive advantage over other insurers.

Because there is little data currently available to provide a robust rating structure, estimates are needed. If there are two components to coverage (physical damage and liability), they need to be rated separately. The replacement costs of drones will vary significantly as well as the accessories. If we assume that any claim would be a full loss, we can estimate the physical damage loss cost as follows:

Physical Damage loss cost=frequency X replacement cost

For liability coverage, we would use a similar formula, changing replacement cost for the expected average claim cost. For higher limits, one may consider a risk load as well (the expected cost of a really high limit may be low, but the insurer would need an incentive to provide this coverage).

Liability loss cost= frequency X average claim cost+risk load

The current data available to estimate the frequency and average claim cost is not credible. Additional coverages such as for libel/slander/cyber risks would also need to be considered. However, if we estimate the average premium for an individual drone to be $1,000 and the expected number of drones to be 30,000 (as projected by the FAA), we can estimate the total premium for the drone industry at $30 million by 2030. However, using the higher projected number of drones provided by AUVSI (160,000 of drone sales in 2025 alone), the estimated premium increases to over $160 million. This amount could increase substantially due to higher limits or higher-value drones and/or accessories. By comparison, the estimated amount of total premium related to the aviation insurance is currently $5 billion, according to Global Aerospace.

Many larger firms such as Amazon may self-insure a lot of this risk and purchase excess insurance in the traditional aviation insurance marketplace. Many large companies already have separate aviation insurance policies to cover either their own corporate jets or fractional ownerships. Drone coverage could be added.

Are there concerns?

While there is a great movement by commercial interests (including insurers) to prime the runway for increased drone usage, there are still many concerns that need to be addressed. Chief among these concerns are issues related to individual privacy. Drones have the capability to collect massive amounts of data—particularly personal images. It is critical to understand and secure the process used by drones for collecting, storing, and deleting data and information. The Florida legislature recently sent a bill to the state governor restricting drones used to photograph or record images of people or their property from the air without their permission. The bill does not include any criminal penalties, but does allow people to sue anyone who photographs them with a drone.

Other concerns include cyber risks related to the potential hacking of drones to steal the data captured by them or to take control of their operations, resulting in a potential safety and security risk. Heavy reliance on digital capabilities make drones appear to be particularly vulnerable to these cyber risks.


Despite such concerns, there is no doubt that drones will have a material impact on the insurance industry in the near future. Drones will affect both the operations of insurers and will provide a sizable new market for products and services. There is much work to be done—as insurers turn art into science.