The Race to Find the Next Insurance Credit Score (or How, Maybe, to Reinvent P/C Insurance Pricing)

The Race to Find the Next Insurance Credit Score (or How, Maybe, to Reinvent P/C Insurance Pricing)

What is an insurance credit score? Basically it is a set of algorithms applied to data from credit reports which provide guidance for pricing and underwriting personal lines insurance. Although it has been a source of political and regulatory controversy over the years, the use of insurance credit scores is now widespread.

Much of the controversy has been over possible disparate impacts on various societal groups. But a root cause of the controversy has been the non-intuitive relationship between a given person’s use or misuse of credit on the one hand—and that person’s probability of incurring insured losses on the other hand. It just doesn’t seem to make much sense. But statistically there are correlations, which in general have passed regulatory review.

Insurance credit score controversies now ancient history (i.e. were settled before most millennials graduated from high school).

But suddenly something interesting is happening.

The race is on to find the next insurance credit score—and the winners (if there are winners) will gain a pricing (and underwriting) edge.

There are only two requirements to enter in this race.

  1. You have to forget about all the kinds of data and information that insurers have been using to price and underwrite risks.
  2. You have to use your digital imagination to find some new data and models which provide the same or better lift as the old data and models which you have just thrown out the window. (Lift is the increase in the ability of a new pricing model to distinguish between good and bad risks when compared to an existing pricing model.)

So what kind of new data might a digital imagination look at?

  • For personal auto, connected cars will provide a rich data set to mine. How about whether a car is serviced at the manufacturer’s suggested intervals (correlated with whether the car is serviced by a dealer or by an independent repair shop)? Or the use of a mobile phone while the car is in motion (correlated with time of day, precipitation, and whether satellite radio is also playing)? Or use of headlights during daylight hours (correlated with the frequency of manually shifting gears in a vehicle with an automatic transmission).
  • For homeowners insurance, connected homes could supply all types of new data. For example, whether Alexa (or other IPA) controls the home’s HVAC systems, correlated with setting security alarms before 11 pm). Or, electricity and gas consumption, correlated with use of video streaming services on week nights. Or the number and type of connected appliances, correlated with the number of functioning smoke, CO, and moisture detectors.
  • For commercial liability insurance, telematics and IoT will be the key data sources. Does a business with 10 or more commercial vehicles use both fleet management and telematics solutions? What mobile payment options are offered (correlated with dynamic pricing capabilities)? The business’ use of social media and messaging apps, correlated with the degree of supply chain digitization.

Of course obtaining a lot of this data will require permission from policyholders—and even with permission these methods may raise social or political issues. But premium discount and loss control incentives for telematics programs have proven effective. And for better or worse, Scott McNealy got it right in 1999.

The Great Insurance Experiment

The Great Insurance Experiment

There is a battle going on today for the future of the insurance industry. Like other industries there are those within the insurance industry and new entrants who are seeking to test whether alternate, digital models will prevail. As a participant in the industry and an observer the intriguing thing for me is no one has proven the existing model is actually broken or that there is a better proposition out there. It seems the telematics experiment I wrote about a few years ago is expanding in focus.

I'm sure taxi drivers said the same when faced with Uber, hotels with AirBnB, the print industry, the travel industry, etc. However let's look at the benefits of digital propositions to customers and see if they apply to insurance.

Transparency
One of the key benefits of digital propositions is transparency and low prices – something that telematics and IoT propositions endeavour to deliver for consumers. The peculiar thing about insurance is that transparency and too much data is at odds with what insurance tries to achieve. Put another way, insurance is designed to hedge the risks to a population across the whole population, so that individuals pay a reasonable price and those that suffer a significant loss are reimbursed disproportionally to what they put in. Absolute data and visibility – transparency in its purest form – will reveal the poor risks and in practice deprive them of the very service they need. Good for some who will not see a loss, but not good for all and not good for society as a whole.

Propositions in this area have moved towards education and rewarding behaviours that reduce risk – the win-win for insurer and client. Many have observed that this is arguably not insurance but rather risk advice, engineering and management. Others observe that claims prevention is absolutely part of insurance and has been all along, albeit the tools of old have been regulation, law and classical education rather than the digital variants.

Existing experiments reveal customers care do care about not claiming, about limiting the impacts of a claim and about small rewards for good behaviour. Regulators have also shown they're keen that all parts of society have access to financial services and insurance at a reasonable cost. Use of transparency and data can go so far in insurance but there are limits to how far it can disrupt.

Control
Another key benefit of digital propositions is the just in time and just enough nature of them – the ability to finely control the product and as a result the costs. This is another area that is being tested in insurance with micro control over what is and isn't on cover available to customers via their phone.

The challenge here of course is that this again removes some of the hedging. By assigning a cost per item turning everything on will typically yield a higher price for insurance than a classic contents policy which offers blanket cover for items in a property or even while travelling.

The other benefit of the classic policy is that one doesn't have to engage with it. It's all well and good that one can turn cover for items off and on quickly but to really take advantage of this capability the insured has to care deeply about the level of cover or the cost.

There will be customers who want this level of control in their insurance and will actively seek it – but for the mass market a good enough policy at a reasonable price will be just fine.

The long tail
Now here we could see some disruption, or at least shake up of the market. We're already seeing some splits in the market as people interested in health rewards take up the various incarnations of vitality insurance, young people take up telematics car insurance after being priced out of the classic policies. There will be customers interested in control over their policies, customers who give up human interaction in favour of digital cost control.

In this way we might see smaller, more agile companies with lower cost bases taking their share of the market by satisfying a niche.

Conclusion
In practice, the jury is still out and the experiment still continuing. Do todays consumers want the products they have always been offered or something new? What of tomorrows customers?

Predictions of Christmas past

Predictions of Christmas past
The speed of technology change is presently both amazingly fast and disappointingly slow. This paradox arises from seemingly huge shifts in technology regularly occurring over the last decade and a half but slow realisation of these in industry. Of late I have personally felt that things aren’t moving quite as quickly as I expected. Since we’re at the end of the year and the holidays are a great to reflect and review how things have gone I thought it worth going back a little and looking at some of Celent’s predictions from 2012. The image below summarises some of the predictions Celent used to highlight just how much change could occur in the following eight years. How much of it has proven to be accurate? Celent Predictions Printing human organs with a 3D printer was a topic of active research in 2011 and the topic of a TED talk. Still a topic of active research and still some years (possibly decades) until we’ll be getting replacement printed hearts and ears. That said, doctors in the US did save a two year olds life with a man made windpipe in 2012. In this case the ambiguous commercial space flight referred to the then-likely space tourism although the efforts of SpaceX have pre-empted the space tourism industry by some years. SpaceX was the first private company to complete a delivery to the International Space Station in May 2012 and made a delivery beyond Earth’s orbit in 2015. Widespread use of 3D printing was another suitably ambiguous phrase. In 2015 every home certainly doesn’t have a 3D printer although the devices are widely used in prototyping activity and are regularly found in increasingly popular innovation labs. The price of 3D printers is coming down to the level where other devices such as home printers and microwaves started to become popular – but the killer application is perhaps missing. Social commerce referred to the seemingly inevitable integration of retail directly into popular social platforms. While retail websites have adopted social features Facebook has not surpassed Amazon in terms of retail, indeed the leading social networks are still advertising platforms and not retail platforms, despite rumours over the last 3 years social networks still don’t have payments integrated in. A prediction firmly not realised. As regards the battery technology the insta-charge batteries are still not here, whilst they are an area of active research. Similarly the idea of highways capable of charging electric cars via induction is still at concept stage – with the adoption of electric cars having been slower than some expected with the popularity hybrids. There’s still time for these predictions to come about but they feel more like a bet than a certainty now. As regards drones executing simple tasks this is already being widely discussed, regulated and piloted in multiple countries. The concept of pizza deliveries by drone – a particular favourite of mine, has already been piloted in multiple cities. Smart energy meters and grids was an early expression of the Internet of Things technology beyond telematics in cars. This is increasingly finding its way into mature markets with multiple insurers in both the US and Europe offering insurance based on devices in the home. Finally, the crash proof car – the topic of Donald Light’s report on the end of auto insurance. It felt far too early to say driverless cars would be ubiquitous by 2022 so this was a safer bet. While it’s a strong statement to say a car is crash proof we have already seen the rise of testing of autonomous cars as well as multiple car manufacturers underwriting the activities of their vehicles while in autonomous mode. We are already seeing manufacturers literally willing to bet their vehicles won’t be responsible for crashes on todays roads. Perhaps then, things are moving swiftly and Celent’s wild predictions of 2012 aren’t that far from the mark. Also of comfort to me is how members of the insurance are directly involved in some of these initiatives, where they are relevant. If you get time to think back on the year, I would be curious on your views. Has technology change sped up? Slowed down? Surprised? Disappointed? Where do you think it will head next? Celent has it’s thinking cap on already and some of these topics will be discussed in our events, What if…. and Celent’s 2016 Innovation and Insight Day, although we’d surely love to hear your views.

Innovation is Magic….Or is It?

Innovation is Magic….Or is It?
What magic does an insurer need to keep up with all the change occurring in our industry? Every firm we talk with is aware of the many challenges currently faced….telematics, digital, social networking, predictive modeling, etc. And yet, more changes are on the way. Our short list includes the Internet of Things, peer-to-peer risk pooling, extended lifespans (maybe never-ending), and artificial/machine intelligence. What should an organization do now to respond and to prepare? What magic is necessary to make it all happen? Celent is pleased to announce our seminar designed to provide answers to how insurers move forward with innovation on practical terms. It will be in London, on 03 February, 2016. The venue, appropriately, will be The Magic Circle, Centre for the Magic Arts. Through a combination of presentation and hands-on workshops, the session will provide attendees with:
  • a practical understanding of how the very basic assumptions underlying traditional insurance products are changing and what impacts this will have
  • firsthand information of how insurers can respond, gained through a series of experiential, structured exercises
  • networking opportunities with peers which allow for comparisons with like organizations
Our view is that innovation, like magic, requires significant work. Successful “tricks” are the result of the investment of time and resources in various techniques, finding what works, what doesn’t and making appropriate adjustments along the way. After a solution is worked out, practice perfects the approach. Eventually, just like magicians during a performance, implementation must be agile, flexible, and respond to changing conditions. As the programme builds, we will post updates. For now, please save the date in your diary and register at this address: https://www.regonline.com/builder/site/default.aspx?EventID=1762713 …oh, and bring your wand!

What happens when auto manufacturers stop giving away valuable telematics data for free?

What happens when auto manufacturers stop giving away valuable telematics data for free?
Here’s a thought experiment. Imagine that you manufacture some things – let’s call them automobiles. And imagine that in those automobiles you’ve installed bunches of computer systems to control steering, braking, transmission, engine performance, and even record GPS-determined locations. Let’s call these computer systems electronic control units (ECUs). And imagine that these ECUs generate streams of data that are potentially highly valuable to organizations that are interested in how safely a given automobile is being operated. Let’s call these organizations insurance companies. And imagine that one day, a really smart person at an insurance company had the great idea that if they could capture and analyze these streams of data, they could understand automobile risks in a way that would let them price and underwrite auto insurance in a much more accurate and profitable way. Let’s call that person Flo. And let’s say that Flo realized that the automobile manufacturers had kindly provided a little port thingy that allows her to access all this valuable and data and transmit it to her insurance company without paying the automobile manufacturers a single penny! Let’s call these port thingys OBD-IIs. And let’s say that Flo and her counterparts at lot of other auto insurance companies go a little crazy giving their policyholders little whats-its that plug into the OBD-II thingys. Let’s call the whats-its dongles. But the really great thing is that the automobile manufacturers are still not charging Flo and her peers a single penny. And let’s say that the automobile manufacturers, one day decide that this internet mobility thing is here to stay, and that it could be a really great way to deliver more value, and deepen their relationships with the people who buy their cars. And to do all this stuff, the automobile manufacturers are going to make cars that connect to the internet! Let’s call these kinds of cars, connected cars. And lastly someone at an automobile manufacturer says, “Oh Dear Dearborn” or “Oh Cool Cupertino” “We could make a bundle of cash by taking a big slice out of the increased profit margin that Flo and her friends have created by charging them very large fees to get access to the ECU data. Or better yet, we could hire some actuaries and data scientists and enter (or re-enter) into the auto insurance business ourselves—and Flo can go make a big bet on smartphone-based telematics.” Ok, so here’s the thought experiment. If your were an investor, named Warren, looking for a growth stock, would you invest in an auto insurance company?

Insurance IoT – you can sense the disruption: Innovation Roundtable summary

Insurance IoT – you can sense the disruption: Innovation Roundtable summary
We held another in our series of Innovation Roundtables in NYC last Friday. These are small gatherings, attended by insurers and banks, meant to provide an open forum for a deep discussion of a chosen topic. As Mick Simonelli, one of the facilitators put it: “The format provides a chance for innovators to come out of their day-to-day battles, take a tactical pause, gain some perspective and share their knowledge with other practitioners.” This edition was focused on the Internet of Things in insurance. More than any other previous Roundtable, the threat of disruption amongst the group was very prominent. The discussion was best summarized by one participant: “We have been doing the business of insurance according to “effects analysis” for 300 years. That is, actuaries and underwriters have been looking backwards and projecting what contract terms (rates, guidelines, etc.) should be going forward. IoT in insurance will provide new territory, which is as yet unclaimed by any provider. It will allow insurance to move toward “causal analysis”: what are the true causes of loss and in what interventions can be undertaken to avoid them?” Much of the discussion was about how insurance risk professionals can accomplish casual analysis using IoT tools and techniques. However, there was also a recognition that other entities, outside insurance, may figure this new approach out before insurers. This may be the well-know data firms such as Google, Amazon, etc., or may be a group of data scientists yet unknown. Other main points from the session include:
  • These practitioners report that they sense that the velocity of change around IoT is different than what has been seen before. Unlike other changes in insurance, decision-makers cannot wait for the data to roll in and the “case to be proved” or it will be too late to respond. Companies reported that they have lost partnership deals with start-up firms because they were unable to make a decision in a timely manner. This dynamic supports the need for a “dual governance track” that has been reflected in Celent’s innovation research.
  • Donald Light presented Celent’s model of IoT, and the group engaged in a good deal of discussion about what part of that ecosystem insurers will want to “own”. There was recognition of the incredible predictive value of the data that will be produced by IoT. However, it was pointed out that what has happened in commercial lines fleet IoT applications is that insureds prefer to retain control over the data as the value for them of using it to manage their fleet vastly outweighs any premium discount that might be awarded. It remains to be seen if this will play out in other lines of business.
  • Regarding commercial lines applications of IoT, it was notable that the group spent as much or more time discussing these opportunities as it did discussing the usual suspects – auto telematics, connected home, and health/lifestyle. Celent sees this as a further maturation of IoT in insurance. The consensus of the group was that commercial IoT is not yet widely addressed and is beginning to be a focus for their companies going forward.
  • A lack of cross-industry integration standards was recognized as a significant barrier to expansion. The participants expressed that there is a need (and opportunity) for a data standards group to facilitate this between insurers and potential device providers. Without such agreement, progress will be more expensive and will take longer than it should.
A final discussion point was perhaps the most exciting. The group is tracking the manner in which IoT is changing the profile of the skills required in insurance. Actuarial science is giving way to data science as more predictive techniques and more non-traditional data sources are used. The participants discussed forming a consortium of insurers to partner with NYC-area universities to establish an insurance data scientist training program. Stay tuned!

CCC acquires a telematics platform: the story behind the story

CCC acquires a telematics platform: the story behind the story
Today CCC announced the acquisition of DriveFactor which provides a device independent platform for telematics data and analysis. Why is an auto physical damage estimation and analytics firm acquiring a telematics platform provider? Well, you could say that telematics is hot, and all personal and commercial insurers writing auto insurance are scrambling to build market share. That is true enough. You could also say that telematics data is going to be increasingly valuable in determining causation and relative responsibility for auto accidents: how fast was each car driving around a corner, who braked first, and how hard, etc.? That is also true. But that is not the whole story. The year is 2018. I am driving a new car, and I am in an accident. My car knows it is damaged. It also knows which systems and parts were damaged and need repair or replacement. And it knows the fastest, best, and least expensive way for that to happen. It might even know about the probability of personal injuries among my car’s occupants. This is all pretty valuable information. Who is my car going to give this information to?
  • Its manufacturer? My insurance company?  Emergency responders? Towing companies? Auto repair facilities? Or software companies that estimate the cost of repairs?
And what is the value of such data from tens or hundreds of thousands of accidents? A telematics platform that can order the flow of this information suddenly starts looking quite valuable.

The other auto insurance telematics shoe drops: who wants to be adverse selection lunch for Progressive?

The other auto insurance telematics shoe drops: who wants to be adverse selection lunch for Progressive?
Until now US insurers have intentionally restricted the impact of their telematics programs by holding riskier drivers harmless. In other words, insurers told policyholders in their telematics programs that their premium could only go down or remain the same. Higher risk drivers’ premium would not be increased even if the telematics device revealed driving behavior which actually deserved a higher premium. But now the world has changed. In its 2014 annual report US telematics leader Progressive dropped this bombshell: “. . . we are affording more customers discounts for their good driving behavior while for the first time, increasing rates for a small number of drivers whose driving behavior justifies such rates” (Celent emphasis). See Bloomberg for the full story. Progressive is saying that when its telematics data indicates a higher premium for a given policyholder, it will charge that higher premium. If that policyholder can find a lower premium at another insurer, Progressive is quite happy to have that other insurer issue that policy, leading (on average) to higher losses, for a lower premium. In insurance this is known as the other insurer experiencing adverse selection. At its most basic level, being a successful insurance company is simple. Understand the risks that are submitted to your underwriters, and charge the right premium for those risks. Progressive is not a stupid company. With this announcement Progressive is signaling that its Snapshot telematics program lets it charge a more accurate and higher premium to certain risky drivers—and it jolly well will do it. If other leading auto insurers’ telematics data leads to the same conclusion, they will have to follow Progressive’s lead. Eat or be eaten.

Personalization in car insurance is just around the corner

Personalization in car insurance is just around the corner
In a 2013 survey in Latin America we asked CIOs about their views on the use of Telematics and UBI. We wanted to know if these were currently in their plans and if they believed that they would have any use and impact by 2016. Despite UBI and the use of Telematics has been well received in UK, Canada and the USA, we found that Latin American insurers were not so optimistic about it. Very, but very, few were in the process of investigating it or considering running a pilot, and the vast majority (overwhelming) said they were not doing (and would do) anything about it in the 3 year timeframe. In our conversations with insurers some would be very cautious about how it could be introduced, particularly on how attractive this would be for a producer to sell. Particularly, successful incumbents reacted as if this had no chance to succeed and that they would not be the ones to try it, as they believed it would negatively impact their current portfolio (lose customers). Others would say that in Latin American countries the burglary component of the premium is significant, while the collision component not so much, and therefor UBI would not bring advantages in price to customers. All these typical reactions from incumbents to an innovative and disruptive idea that provides the means to personalize the rate to reflect the real risk that the driver represents. Pay as you drive, pay how you drive and more lately manage how you drive are value propositions that target to personalization (of risk) and loss prevention, two of the major trends we see in insurance in the future. As any disruptive initiative, it only takes one to be bold enough and then change the dynamic of the market. We were conscious about some limited amount of initiatives that were being considered during 2013 in the region and our position was that the 2013 survey results would change completely as soon as UBI and Telematics was taken seriously in the region by at least one player (regardless of the size). We also kept thinking about leading incumbents. Good for them if this did not succeed. But what would happen if they started losing the good risks towards an insurer with an UBI value proposition? For sure their revenue would be affected. But then, how would the leading incumbent’s portfolio look like if only the high risk drivers stayed? Not a pretty scenario, ha? Some interesting facts that occurred since last year. The few pilots are taking shape; research is also indicating that in fact specific segments of customers pay more for car insurance just for being younger or a combination of factors that have nothing to do with the real individual risk; the use of telematics has seen its first implementation in a producer distribution model, moving away from being exclusively a direct insurance proposition (or targeted by a few specialized brokers); and last week Baseline Telematics, a leading provider of telematics based solutions, and Sistran, a leading provider of core insurance solutions, made available to Latin American insurers a combined offering with all the required technology for insurers to quote, sell and price insurance policies entirely based on the actual driving habits of a driver (mileage and behind-the-wheel behavior), which is obtained through a telematics device installed in the driver’s vehicle. Guess what? Not surprisingly our 2014 survey (in edition) shows that, as we anticipated, the perspective on UBI and Telematics has changed completely in Latin America! Around half of the respondents indicated that they believe that in 2014 these technology will have some kind of impact or at least will be tried as a pilot, but most importantly, the view 3 years from now (2017) is that the majority believe it will be of use and impact in underwriting, rating and claims (against only 20% that indicate no use expected). Solutions as the one presented last week enables monthly billing, with a variable premium based entirely on the usage of the vehicle (kms/miles) as well as the behavior of the driver (acceleration, braking and excessive speed). The objective is to attract the best drivers and rehabilitate (or eventually get rid of?) the more risky ones. The goal is to decrease exposure to risk (cost reduction), perfect your technical margins, and gain market share. It seems now that the market has aligned in the right trajectory and personalization in car insurance is just around the corner. Will leading incumbents take their chances? Will this be the opportunity for others to grow in market share and quality of their car insurance portfolio? Does any of this resound familiar to you in any given market you may be? If you are interested in the topic, feel free to contact us. Also, these are some of our reports related to this subject:

Creative Disruption in Action: Changing the insurance outlook for young drivers

Creative Disruption in Action:  Changing the insurance outlook for young drivers

With less than a month to go until our Creative Disruption Symposium in New York on 13th September 2012, the Celent team are working hard on pulling together some great content. The agenda and speakers have been confirmed, and hopefully we’ve got the technology lined up to add in a bit of audience participation for fun.

On Monday, I spent a great day with one of the new breed of telematics insurers in the UK that base their model on ‘Pay How You Drive’, called ingenie. We’ll be featuring them at our symposium in a video. What amazed me about this start-up was the passion and energy not just around delivering the insurance product through new technology but also the desire to change the driving behaviour of young 17-25 year old drivers. This passion extends to bringing new disciplines into the risk pricing equation, such as behavioural science to understand young drivers’ attitude to risk and also top-end driving science through a partnership that they have developed with the Formula 1 Williams team. Based upon their discoveries, they have added ~300 algorithms into the risk selection and pricing equation.

They also use the same information to feedback driving performance to young drivers in a way that they want to receive it, via a combination of a mobile app and push notifications. There’s no point in pushing the data out to a traditional browser based portal as that’s no longer how 17-25 year olds want to interact with technology. The goal of this model is to influence behaviour in order to reduce the total claims cost, build a long-term affinity with the young driver, and in doing so deliver a stable return.

For years, the traditional UK auto insurance industry has dismissed young drivers as virtually uninsurable, backed up a claims experience that’s hard to argue against. And it’s no surprise that this is the response when you consider that the traditional model has delivered an above 100% COR for entire UK auto insurance market over many years. In 2011 alone, which was considered to be an improvement on prior years, the industry made an operating loss of £600m ($960m).

Through changing the model to focus on adapting driving behaviours and in doing so reducing the frequency and cost of claims, telematics is enabling new entrants to target this underserved market using a viable alternative capable of outperforming the industry incumbents. As a result, it’s no surprise that many of the insurers that we speak to in the UK are seriously looking at how (or even if) they should respond.

To us, this is a great example of Creative Disruption in Action and one that we will be covering in more detail at the Symposium. There’s still time to register!

Finally, whether you are able to attend our Symposium or not, why not help us prepare by taking five minutes to complete our survey on Creative Disruption within Insurance. Click here to participate. Thank you.