A new direction for Marine Insurance

In the 1830’s two ships a day were being sunk off the coast of the British Isles. Fraud was endemic. Early marine underwriters only survived if they had access to the best information. Networks of agents around the world helped ensure that insurance wasn’t provided to a ship that had already sunk.

Today less than 100 ships a year are lost globally. Frequency may have reduced, but severity is increasing. The last five years has seen the largest vessel loss (Costa Concordia, $2bn in 2012) and the largest loss in the marine market (Tianjin explosion, $6bn in 2015).  Yet marine underwriting isn’t much different today than it was when ships had sails. Could this be about to change?

Crew negligence is by far the most common factor in causing ships to founder. Regulations on the operation of ships and standard of crew vary from country to country and are difficult to enforce.  The market is competitive. Profit margins are shrinking.  Poorly trained crews, mis-declared cargo, overweight containers, poor maintenance, sailing in restricted areas and contaminated fuel are hard to track and becoming increasingly common as operators look to cut costs. Meanwhile ship sizes are increasing. New “mega-ships” create the potential for larger losses from expensive cargos, complicated salvage operations and third party liabilities.

Most marine underwriters are keenly aware of the need for better analytics but struggle to find data of sufficient resolution and accuracy to use when selecting risks and setting prices. Many use the AIS (Automatic Identification System) data.  Mandatory for all ships over 300 gross tonnes, this is now used on almost all commercial vessels. AIS gives the position and past movements of individual ships.  AIS data is available for free and provides some insights to underwriters on a ship’s location, but has many limitations.

Until recently there has been a lack of tools to enable an underwriter to select and monitor vessels and build a high performing, well diversified insurance portfolio. Indeed, there appears to be an inverse correlation in the insurance industry between the amount of data available to underwriters and the value of the risk.

Over the last 20 years we have become used to providing detailed information to obtain car insurance. Driving history, claims, engine size, location and driver’s age all materially affect our premium. Insurers combine what we tell them with comprehensive independent third party data and their own actuarial calculations to derive a rate that accurately reflects the risk. Insurtech start-up companies are offering mobile phone insurance priced using detailed data on the model, value and even location and frequency of use of the phone all automatically collected by the app used to buy the insurance. Marine underwriters can only look on in envy.

This could be about to change. A consortium including Maersk, Guardtime, EY, insurers MS Amlin, XL Catlin, brokers Willis Towers Watson, standards agency  ACORD and Microsoft announced in September that they are building a marine insurance platform using blockchain. Details are still sketchy, but those involved say that the lack of established technology for marine underwriters made this an attractive proposition to test out the latest developments in blockchain and distributed ledgers.

Maersk’s involvement is also significant. The Danish business conglomerate is the largest container ship and supply vessel operator in the world, valued at over US$35 billion. It’s rare to find a large company partnering with an insurer. Risk managers, the buyers of corporate insurance are conservative. They rely on their broker to advise on insurance cover and generally opt for conventional insurance structures and covers. Many of the new and well run ships have sophisticated sensors monitoring the performance of engines and other key systems. If partnerships can be developed directly with ship owners and fleet managers that gives access to this data, combined with other third party information, there is the potential to create mutually beneficially and accurately priced insurance coverage.

Sam Evans of insurtech investors, Eos Partners, identified marine insurance as a key area of focus at the start of this year. Sam noted Concirrus’ announcement of the release of Quest, its marine underwriting platform, along with the Maersk blockchain platform as two major developments that will impact this market.  

With insurance premiums of $30bn the marine insurance market is over 10 times bigger than cyber insurance. Underwriters are increasingly motivated to improve how they are selecting and underwriting hull, cargo and the associated liabilities that make up this market but they will not make changes lightly. Insurers want to manage their loss potential, but as with most lines, there is an excess of capacity in marine. For underwriters to grow their premium income they need to be offering rates below their competitors.

As Sam Evans notes data and technology providers to the marine industry have been fragmented, with no dominant vendors and no joined up, end-to-end solutions. The market is big but not huge. The winners will be the technology companies that give the underwriters confidence to rethink how they set their rates. They need a range data sets that are credible, current, reliable and comprehensive, served up through a sophisticated analytical platform embedded in their workflow. The marine market may have not changed for a long time, but that doesn’t mean that it can’t change quickly.