Technology is disrupting risk, and it will both change and make maritime fortunes in the process.
When Apple’s new iWatch went on sale recently the consensus from commentators seemed to be that whilst it was very nice, it just wasn’t something anyone really needed. But the iWatch is already proving very useful indeed, and not for telling the time.
In early April Manulife Financial Corp.’s John Hancock Life Insurance announced that it would discount premiums by as much as 15 per cent for policy holders who tracked and performed well on metrics such as daily exercise, annual health screenings and flu shots. “We want to make life insurance more immediate and relevant in the daily lives of our policyholders and help them connect their financial well-being to their long-term health,” John Hancock President Michael Doughty said in a statement.
But Hancock wasn’t going to blithely hand out discounts to customers on the basis of their own activity reporting. The deal came with a Fitbit wearable fitness tracker which monitored every step its customers took. A month down the line with the launch of the Apple watch the insurer has already moved from a mere step tracker to an app which allows customers to use the Apple Watch, iPhone and iPod Touch to meet the health-tracking requirements of the insurance policy.
The app allows policyholders to post and view their points and status, earn points by submitting prevention screenings and athletic events, and log health club workouts. It also integrates with Apple’s HealthKit so that participants can complete their health review via the app, which itself gathers data from the Apple Watch—including real-time steps.
The whole idea of your insurance company having intimate, near real-time knowledge of your exercise and activity habits might make you very uncomfortable. But if it does then the chances are you’re not a Millennial, or one of the subsequent generational cohort, Gen Z.
For those generations the idea that data—and your own in particular—is a currency you can use to purchase goods and services either discounted or for free, is already deeply embedded. The combination of this digital native mindset, connectivity, and advances in a variety of technologies are colliding to fundamentally reinvent the way that we evaluate and manage risk, and the consequences will be wide-ranging for individuals and businesses, on land and at sea.
In 1960 a movie called The Apartment was released. It stars Jack Lemmon as CC Baxter who opens the film by telling us that he works on the 19th floor, Ordinary Policy Department, Premium Accounting Division, Section W, Desk number 861, for an insurance company called Consolidated Life of New York. Baxter’s character is obsessed with the kind of statistics which have traditionally allowed actuaries to analyse and understand the cost of risk, and quotes them frequently.
But whilst these statistics have provided the bedrock of the practice of insurance, they have always been broad and often blunt tools. The shortcomings of Baxter’s statistics are pointed out to him by the girl he’s in love with. She tells him she never catches colds to which Baxter replies that according to figures from the Sickness Accident Claims Division the average New Yorker between the ages of twenty and fifty has two and a half colds a year. “That makes me feel just terrible,” she responds. “To make the figures come out even, if I have no colds a year, some poor slob must have five colds a year.”
In essence everyone who’s ever been insured to date has been paying for the poor slob. But in the past 10 to 15 years advances in computing power and the proliferation of new digital data sources means that insurers are moving beyond the core actuarial disciplines of advanced maths and financial theory and opening up a new paradigm.
The core data for insurers used to be internal—claims histories for example. But third-party data sources have expanded massively, with previously inaccessible public-sector data now available and both the EU, UK and US governments launching open data websites containing huge datasets of statistics around health, education, safety, energy usage and much more. Add to that the constant streams of ‘exhaust data’ from social media, multimedia, smartphones, computers and other consumer and industrial devices, and the landscape of behavioural insight alters completely.
But as Hancock’s policy shows, advances in our technological ability to monitor and visualise in real-time may disrupt the way we approach risk the most. Hancock is in health, but in car insurance the trend towards Usage Based Insurance (UBI) and telematics has been steadily growing. Using increasingly small and sophisticated devices inside the car insurers are monitoring the real-time driving habits of their policyholders. That means that instead of you paying for what the poor slob might do, you’re saving money based on what you’re actually doing.
It is a fundamental shift both in the way risk is evaluated and also of the relationship between us and those who insure us. And what’s particularly interesting is the influence of this new, real-time relationship on driving habits. According to McKinsey one UK insurer offering telematics-based policies has demonstrated that monitoring leads to better driving which has led directly to a 30 per cent reduction in the number of claims. Another reports having used telematics to enable a large client to reduce accident-causing risky manoeuvres by a stunning 53 per cent.
“We often speak to companies who are reinsuring ships where they don’t know who the operators are and they certainly don’t know what cargo is being carried,” Adam Szubin, U.S Treasury Department
The underlying mega-trend here is what’s been called ‘perfect knowledge’. Connectivity, the Internet of Everything, advances in computing power, machine learning, algorithms, the falling cost of microprocessors and the ability of technology like GE’s Direct Write to ink sensors on the parts other technology cannot reach, means we are creating a world where in future we can know everything about anything immediately and accurately. At least, that’s the idea.
As the sci-fi writer William Gibson pointed out, the future is here, it’s just not very evenly distributed. And one place where that future is particularly patchy is the blue domain.
“Man marks the earth with ruin; his control stops with the shore,” wrote Lord Byron in his poem The Ocean in the early nineteenth century. And despite the advent of deep sea satellite communications in the twentieth century, that control—and crucially knowledge—to a large extent still stopped at the shore until comparatively recently.
I’ve written in depth before about how profound and fundamental an enabler connectivity at sea has become. (Read our connectivity issue of July 2014). The proliferation of communications technologies from HTS systems like Intelsat’s EPIC and Inmarsat’s GX to nanosats from companies like Google, Spire and Space X are seeing the future distributed far more evenly across the oceans than ever before.
The lesson is that digital operation and connectivity breeds transparency and that’s exactly what’s happening in maritime right now. The flows of what I describe as Blue Data are vast and increasing, and they are offering the opportunity of unprecedented insight, knowledge and advantage. The future is spreading fast.
The early-adopters seizing the opportunity however, aren’t shipping and maritime companies per se. Those leveraging the insight this data can give them are different stakeholders. But when it comes to transparency they’ve got considerable skin in the game.
Following the financial crash banks and other institutions have been subject to an increased level of scrutiny, and not just around their financial health and viability. Focus has also fallen on other risk, including corruption, money-laundering and compliance with global sanctions. Considering it transports 90 per cent of world trade the shipping industry has remained stubbornly opaque, using its complex ownership and operation structure to defy the kind of transparency commonplace in other industries. But perfect knowledge breeds a new kind of expectation and the new availability of maritime data and insight means that those who make money from the sea aren’t going to be able to use ignorance as a defence in the future.
Earlier this year speaking at an anti-money laundering conference Adam Szubin, director of the U.S. Treasury Department’s Office of Foreign Assets Control said that big potential sanctions risks may lurk within the business operations of securities firms and reinsurers. Omnibus accounts, he went on to tell his audience, which combine the transactions of multiple parties, pose a risk for securities firms if they don’t know whose assets they are holding. “We often speak to companies who are reinsuring ships where they don’t know who the operators are and they certainly don’t know what cargo is being carried,” said Szubin. Pointing out that opaque lines of business have proven to be a big compliance challenge for companies, he had this warning, “I urge you all to think carefully: what in your business lines do you see as those black boxes? That’s typically where the threat is going to come from.”
And the threat to institutions from regulators is also growing as post-financial crisis there has been a new toughness of approach. Under Benjamin Lawsky the relatively obscure New York Department of Financial Services suddenly showed its teeth in 2012 by threatening Standard Chartered Bank—accused of laundering money for countries including Iran—with revocation of its New York banking license without which it could not operate.
With that kind of repercussion on the horizon it’s little wonder that financial institutions, traders and insurers would be first in line to sign up for platforms which leverage blue data flows to deliver transparency and knowledge. And they’re funding them too. As we went to press it was reported that Israeli company Windward had secured $10.8m in investment from Horizons Ventures amongst others in order to, “accelerate Windward’s ability to build the largest, most comprehensive maritime data and analytics platform in history.”
Both the Trident Alliance and the Maritime Anti-Corruption Network are collaborating to leverage transparency to bring about a level operational playing field. Technology can deliver it.
That’s a big claim from a company you might well never have heard of, but Windward’s intelligence solution, MARINT, is apparently already in wide use by security, intelligence and law enforcement agencies worldwide, who use its data and insights to pre-emptively identify threats before they reach their shores.
Windward has taken datasets including ‘port agent reports and ship bookings’ together with either the past two, or four years worth—depending on which report you read—of AIS data for more than 200,000 ships, plus cargo reports, ownership data and twenty five years of ‘ship economics research’ according to Ami Daniel, Founder and CEO, mashed it up and allowed its algorithm to go to work. The result is individual ship profiles which allow Windward to identify which vessels may warrant further investigation by the authorities.
Windward has also been aggressively denouncing the unreliability of AIS data, putting out a report claiming that ships on average report their final destination only 41 per cent of the time, 55 per cent of ships misrepresent their port of call or nearby port, for most of their voyage, and that over the past two years GPS manipulation has risen sharply by 59 per cent.
Its platform addresses this by, “checking every transmission on the bit and byte level and cross checking identities, geographical movements and economic profiles globally and historically,” Michal Chafets, its Director of Communications is quoted as saying. “Together, this creates a PayPal-like fraud entity resolution and fraud detection technology that vets the AIS data.”
If accurate Windward’s estimate that around one per cent of all global shipping is up to no good and deliberately trying to throw everyone off the scent means it’s little wonder that since launch five years ago Windward has secured government and security services as customers. But as ever stricter IMO mandates begin to bite and stakeholders including banks and cargo owners increase their CSR scrutiny, top tier operators themselves are beginning to demand greater transparency to address their own risk.
Take the Trident Alliance for example, a coalition of owners and operators collaborating to bring about more robust enforcement of sulphur regulations which are raising costs for them and providing those who don’t comply with a cost-advantage. According to the alliance only one of every 1,000 vessels in Europe currently gets tested and of those around half are in violation of the sulphur regulations. The Maritime Anti-Corruption Network (MACN) is another example where operators, cargo owners and suppliers are working together towards the vision of a maritime industry free of corruption that enables fair trade for the benefit of society at large.
Both organisations are focussed on transparency that can provide a level operational playing field. New initiatives like the $12.5m EfficienSea2 project—which aims to develop a range of new digital services including automatic ship emission reporting and monitoring—are an example of how this transparency might be delivered in the future, but that’s part of a far wider ecosystem developing at sea.
The two-tier efficiency market is being positively reinforced. In 2014 three ports began offering financial incentives to efficient vessels.
Pioneering connectivity suppliers like KVH are creating new platforms enabling products and services to be delivered and structured in new ways, and data transmitted and accessed far more widely. At the same time there is a revolution in how some of the most humble products are being approached. It is almost incomprehensible to many outside the industry that until recently the Voyage Data Recorder (VDR) onboard was subject to twelve hour overwrites, but the new standard has formed a jumping-off point for innovative manufacturers like Danelec. Its next-generation product includes features such as playback software for real-time monitoring and replay of recorded data, together with data capture and analysis.
We are rapidly approaching a situation where the availability of maritime domain and ship specific data will be available in real-time to a wide variety of stakeholders, and whilst that should enhance security, safety and efficiency it will also expose operators to a potentially disturbing new reality. Like those policyholders checking their Apple iWatches and desperately jogging their way to a fifteen per cent discount on their life insurance premium, ship operators may discover that the availability and quality of their operational data becomes pivotal.
In the same way that car insurers are taking real-time data from their drivers, marine insurers could sometime soon have access to a rich set of data streams that will allow them—if they’re given access—to know exactly where a ship is and where it’s headed, what it’s carrying, who’s on board, how healthy they are, how much fuel the ship is burning, whether the engines are being operated in compliance with the manufacturers recommendations, whether the systems on board have been appropriately maintained, whether the ship is transmitting the correct identification data, whether its network and systems are secured, whether its emissions are appropriate, whether its charts and software is up to date, whether the crew hold the appropriate tickets and training, whether the spares it’s carrying are authentic and exactly how much that ship is worth on that particular day in any currency they choose.
For those operators for whom operating beyond compliance is the norm, policies based on a transparent usage model could well prove highly attractive, but they could do something else too. By demanding transparency they would not just level the playing field, they could actually tip the balance in favour of responsible operators. The ones everyone wants at sea.
If the experience of this kind of data-led insurance ashore is replicated on vessels then the impact could be significant. In short the opportunity would exist to actually incentivise and reward positive behaviour by ship operators with reduced premiums whilst starting to raise the barriers to entry into the industry for those who currently operate non-compliant or otherwise unsafe ships.
The likelihood is that we would quickly see a two-tier industry develop, with those able to demonstrate a positive dynamic risk profile and those who can’t or won’t. Ultimately your ability to insure your vessel at all might well depend upon the real-time data you’re able to produce.
In fact we’ve already started down that kind of road. A recent survey by the Carbon War Room found that vessel efficiency rankings such as the A to G GHG Emission Rating developed by vetting company RightShip now form an important part of assessing risk and return in several banks, with inefficient vessels now representing a higher-risk investment.
Energy efficiency data is also being used in credit-approval processes for vessel purchases, loan assessments for retrofit projects and re-sell of scrapping decisions, with banks citing efficiency as a key indicator for a vessel’s profitability.
“Banks are now publically stating that they use energy efficiency data in deciding which vessels will receive financing and which ones won’t,” says José Maria Figueres, Chair of the Board for Carbon War Room and Rocky Mountain Institute. “As this trend continues, inefficient ships will become increasingly unmarketable.”
“As a consequence of the correlation of energy efficiency and loan risk we have analysed our shipping portfolio based on the methodology of the EEDI and implemented design efficiency criteria in our credit approval process,” confirms Carsten Wiebers, Global Head of Maritime Industries at KfW IPEX-Bank. “In view of the beneficial risk profile and environmental benefits we favour eco-ships over ships with poorer energy efficiency.”
This incorporation of efficiency data into financing decisions is seen by the CWR at least as an indication of a dramatic market shift in recent years with banks saying they’ve seen the formation of a two-tier market comprising high and low efficiency vessels. Eco efficient vessels demand a premium price at new build stage, are more likely to be chartered, maintain asset value over time and have a longer lifespan. “We see a clear trend towards a two-tier market of high and low efficiency vessels—more energy efficient vessels have an enhanced marketability as well as a higher revenue potential for the ship owner and thus a more favourable risk profile for financiers,” says Figueres.
The two-tier market in efficiency is beginning to be positively reinforced too. In 2014 three ports— Port Metro Vancouver, Port of Prince Rupert and Port of Barbados—began to use the A to G rating to offer financial incentives to the owners of more efficient vessels entering their ports, and now twenty five per cent of the non-container charter market vet potential vessels for efficiency before charter. But that’s just one metric. Where the insurance market could go involves not just the value of the asset but the way it’s operated, day after day.
Simply having an efficient vessel by design is as blunt a risk measure as being of one gender or another. When it comes to your risk you’re still paying for the poor slobs with five colds per year, even if they’re more sustainable colds that burn fewer bunkers.
That data on its own is a start but it only scratches the surface of the rich insight that the vast datasets we’ll soon be able to access about ships and shipping could give us. RightShip has been at the leading edge of just this kind of insight for some time and so it’s no surprise that it’s also leading on potentially the greatest disruptor—prediction.
Speaking recently at a conference in Australia RightShip chief Warwick Norman described how the company is beginning to move from reactionary to predictive vetting and using IBM predictive analytics to help them. In the process it’s turning up some counterintuitive insights, particularly with regard to age versus incidence. Tankers are likely to be far safer than other vessel types, but newer vessels aren’t necessarily the ones with the lowest risk profile.
It’s insights like these that could rewrite the way we insure ships. And that has massive implications for the P&I Clubs. If it’s correct, as one source tells me, that the maritime insurance industry hasn’t made a profit in the last seventeen years, then it isn’t coming a day too soon. Certainly there’s at least one P&I Club actively working on a predictive risk model and rumours of many more similar projects in different stakeholders around that same predictive analytics theme.
I stuck my neck out back in October and identified prediction and autonomy as the two trends which were likely to have the most profound impact on the blue industries, but even I’ve been surprised at how fast that’s happening.
Windward just raised $10.8m to target financial institutions it claims are ‘hungry’ for advanced solutions providing valuable trading opportunities by extracting faint signals from noisy datasets like maritime information.
Windward’s current business may be in maritime domain security but its new funding is designed to support its expansion into different territory, enabling it to expand its deep learning capabilities via its data platform called ‘the Windward Mind’ and to operationalise FORESEA, its finance solution. Currently in beta testing FORESEA is the first significant vertical extension of the Windward Mind and according to the company is providing traders, investors and analysts with access to unprecedented amounts of unstructured data, critical insights and untapped market opportunities.
“The Windward Mind, the world’s first maritime data platform, brings cross-vertical and industry visibility into ship activity worldwide that is critical given the economics at stake,” said Ami Daniel, co-Founder and CEO of Windward. Tom Glocer, the former CEO of Thomson Reuters, who will also be joining Windward’s Advisory Board took up the same theme. “Windward is a game changer because it is bringing a data sciences perspective to today’s maritime data and making it accessible and relevant for the entire ecosystem,” Glocer said in a press release. “Financial institutions, the next market for Windward’s unique data platform, are hungry for advanced data solutions that can provide valuable trading opportunities by extracting faint trading signals from noisy datasets such as maritime information.”
But Windward has some stiff competition when it comes to bringing data science to bear in the blue domain. Warwick Norman at RightShip is leveraging predictive analytics for insurers. This issue’s Futurenaut Richard Rivlin’s company VesselsValue can now predict the long term value of a vessel. Former Futurenaut Paul Østergaard has developed the Shipserv Match engine that predicts which suppliers are the best fit for your procurement needs based on fifteen years of transaction data on the platform. And what about Xeneta, busily disrupting the $200bn sea-freight industry?
“Predictive is such a huge topic but we are talking to data scientists about how to pull in external sources,” says Xeneta CEO and founder Patrik Olstad Berglund. “We have a German based investor and in their portfolio is Risk Methods who have a dataset we could potentially match with ours which is collecting loads of different data streams to notify you about risks in your supply chain.” And whereas Windward may be building a platform, it doesn’t appear on the surface to have the proprietary and extremely valuable datasets the likes of VesselsValue, ShipServ and Xeneta have been growing, on which predictive algorithms can be let loose to generate really game-changing insights.
“Predictive opens up hedging and derivatives,” says Berglund. “I want to get to the point where we don’t report averages any more but that we show the spread—the high and low of a particular carrier—and project what can we expect of its development into the future. Then you can also project/predict that carriers’ revenues, and give that intelligence to banks handing out loans to carriers. Who is the most solid? Who has most contracts moving into the future at the highest average price? All kinds of analytics we can leverage the data for to give access to other stakeholders ultimately aside than those mostly focussed on how it can benefit the buying and selling side of the industry.”
There are a lot of people who really don’t like the idea of an industry exposed to this level of scrutiny, but one suspects that often stems from a real fear about where monitoring, analysis and insight is taking us and what it means for the opaque, murky world some operators thrive in. The reality is though that we need this new transparency and collaboration in order to manage entirely new kinds of risk.
It’s been cited often that autonomous, crewless vessels will be impossible because no one will insure them. But the rise of autonomous systems, whether there are crew on board or not, are already creating new areas of risk and responsibility we have to manage. When software and systems are effectively running the ship where do the lines of responsibility of manufacturer, crew, operator and owner get drawn? Shall we see manufacturers having to cover the risk of collateral damage should their own machinery or software fail on board? As cyber-security becomes ever more pivotal could connectivity suppliers find themselves with more responsibility to end-users of the network?
Crucially, how will the responsibilities of the humans on board ships change as more autonomous systems are introduced? It can’t be long before the operation of the complex systems on board will be beyond the ultimate control of the Master. Is it reasonable then to assign him or her ultimate responsibility?
These questions will need to be answered, and the insight we require in order to do so adequately resides in the massive data flows we’re on the brink of harnessing. And if connectivity is the key to making usage based insurance work then in the same way that car insurers are providing low cost telematics boxes, it may well be that marine insurers realise it’s time they did the same.
The insights they stand to gain could cover everything from crew health and welfare via quantifiable health monitoring, to cyber hygiene, competence, machinery health and the comparative safety of autonomous systems and the human element.
Huge datasets and predictive analytics won’t just enable us to cope with new risks, they will offer new sources of competitive advantage for ship operators too.
That level of deep insight would enable hitherto unimaginable analysis of real-time dynamic risk, and a correspondingly smart insurance industry that reinforced the positive, sustainable operations we’re all calling for. And if the data demonstrated that autonomous, crewless ships were a better risk we could even find insurers taking evidence-based decisions to be in the vanguard of moves to cover them, in preference to ones with crew onboard.
But these huge datasets and the predictive analytics we could bring to bear on them aren’t just important for risk. They could also lead to entirely new sources of competitive advantage for operators. Not only could insights enable shipping companies to identify potential customers far more accurately, in the future they might just allow us to read the market well enough that we can put ships where our customers need them, before they know themselves.
I put the idea to Xeneta’s Berglund, “If you had enough transactional data, you could actually do that. If you had all the booking and all the cargo movements in the world, and we’re collecting contract data, AIS data, so if you knew where cargoes moved you could see patterns in that for certain periods—more cargo from India to NW Europe at certain times of the year—and you could start doing predictive analysis on that. You could predict that in two months time there’s going to be more volume coming out of Vietnam or Thailand and you could get vessels in the right place. I think it’s a fascinating idea.”
We’re living through the last days of the old risk paradigm, the fortune cookie era when the insight we needed was locked away and all too frequently we only understood the risk in retrospect when things got broken. The technologies delivering perfect knowledge and the power of prediction are already causing seismic shifts to the extent that they are disrupting risk itself. They will change, and probably make, fortunes for many in maritime.
But no matter how powerful the stochastic capabilities of our algorithms, we shouldn’t forget that the future isn’t somewhere we go, it’s something we create, and nothing is going to change that anytime soon.
As CC Baxter would say, that’s just the way it crumbles. Cookiewise.
Images courtesy © Getty Images/Apple/Unites Artists/Sibos/Virgin.com/Greenship/Rolls-Royce/Xeneta
This article appeared in the April 2015 issue of Futurenautics.
Download the full April 2015 issue free of charge as a PDF, read online and subcribe