As the insurance industry races to adopt new technologies and stay one step ahead of the insurtech disruptors, blockchain has become a widely discussed topic. With use cases in fraud protection, risk management, claim processing and smart contracts, blockchain has a promising future with benefits for both independent agents and carriers. Although adoption is still in its initial inception, interesting pilot use cases are popping up across the industry.
Blockchain’s greatest value lies in its distributed ledger technology, which acts as a uniform source of truth. This technology is very hard to hack and provides a wealth of benefits to every member of the insurance distribution channel. Let’s take a look at some of the ways blockchain is being used in insurance today.
In my experience and research, the most commonly discussed use case of blockchain lies in the execution of smart contracts. For those unfamiliar with the concept, blockchain’s distributed ledger allows one computer to register an outgoing transaction and then enables a peer group of computers to validate and accept the transaction through a consensus-driven approach. This means that copies of the transactions are now stored in a distributed fashion across the network, and hacking or altering this information will require more than 50% of the computers in the network to be compromised simultaneously. This ensures that the source of truth is preserved and protected in a robust fashion and can be accessed by any legitimate party with the right permission levels.
While the execution of smart contracts in travel insurance has already made a splash – see AXA or FlightDelay or Lemonade – there remain other use cases that are sure to make a large impact. One example is flood insurance. In areas prone to heavy rainfall and flash floods, insurers can use regional geological data to automatically trigger insurance claims. Meaning, once flood waters reach a predetermined level, a smart contract trigger will spontaneously file a claim for the insured. Another application with a similar process is earthquake insurance. If an earthquake were to occur above a certain magnitude, the smart contract would initiate the insured’s claim based on regional geological data and preset factors in the contract.
I also see blockchain expanding into the auto insurance space through the use of telematics devices. These devices are already able to track data in terms of wear and tear, collisions and driving patterns. Such data can be used to calculate insurance premiums that are more targeted and personalized. In a broader sense, IoT-based sensor data can power the metered insurance space in the shared economy (Uberization).
For both the insurer and the insured, the smart contracts built on the blockchain will drive more efficiency across the insurance value chain. A key factor in the expansion and adoption of smart contracts in the insurance industry is data. Smart contracts are only viable if there are external sources of data that are validated and reliable. The more data that is universally shared and available, the more innovative insurance products will be adopted.
Fraud Protection and Proof of Insurance
The FBI estimates the U.S. government spends more than $40 billion per year on insurance fraud, leading to my next, and possibly most compelling, application of blockchain. With an aggregated repository of data that is validated and maintained by carriers, agents and government entities, it becomes easy to track down insurance fraud spanning multiple carriers. IBM has already announced a new framework for securely operating blockchain networks to directly fight insurance fraud, and I expect more companies to follow suit.
In addition to fraud protection, blockchain technology is powering another compliance innovation: proof of insurance. In December 2017, a consortium of insurance leaders dubbed the RiskBlock Alliance launched RiskBlock, a proof-of-insurance tool built on the blockchain framework. It was designed to help insurers, insureds and law enforcement simplify how they verify insurance coverage in real time, eliminating the need for paper-based insurance cards. Nationwide is already in the pilot stage with RiskBlock and hopes to expand the program this year.
Insurance Distribution Model
Despite new insurtech entrants disrupting the industry every day, innovations such as blockchain ensure that the insurance agent will always remain the cornerstone of the insurance distribution model. As reported in the examples above, the common theme among the benefits of blockchain’s distributed ledger technology — the ability to automatically file claims, process data and inform policies — drives efficiency and visibility into the entire insurance ecosystem. For carriers, a uniform and validated data source allows transparency into risk assessment, underwriting and a channel to reach the end-insured directly. With automated processes executed through blockchain, brokers are able to focus on building relationships, expanding their offerings and solidifying their role in the distribution model. Similarly, customers benefit from personalized and increased touch points, leading to better tailored insurance policies and cost efficiencies. As these relationships grow, so does the velocity of business…it’s a win-win for all the constituents in the value chain.
We are only just beginning to see the potential of blockchain technology in insurance. With blockchain and insurtech startups, coalitions such as RiskBlock Alliance and major carriers leading the charge, the insurance industry is poised for an imminent digital revolution.
In the past five years, three technologies have laid the foundation for remaking entire industries:
Advances in IoT technology have burst open the floodgates of real-time data.
AI can process this massive amount of new data to identify previously unknown correlations between inputs on risk.
Distributed ledger technology has obviated the need to either, (to borrow from a Russian proverb), trust or verify data provided by a third party.
Though use of these technologies in insurance applications is still in the early days, it is clear they will have profound impact on the industry. This paper will explore how these three transformative technologies might be woven together to create a single platform, enabling insurers to mitigate claim events, slash operating costs and
improve the customer experience.
Successfully implementing such a platform will require a significant change to the insurance business model. Insurers will expand their role beyond just that of a counterparty to whom risk is transferred and become a critical business partner providing operational, logistical, and business process services to their clients.
IoT and sensor data provide granular data in real time on processes and conditions that were previously detectable only through post-production Q/A processes or manual checks. Distributed ledger technology (DLT) allows the reporting of this data to become immutable. Large amounts of “shenanigan-proof” data can be run through an intelligent rules engine to create smart insurance
There has already been considerable interest in using DLT and IoT data to create smart contracts to insure objects that are the sources of that IoT data. Aigang has a proof-of-concept application for insuring mobile phone batteries.
The transformative opportunity lies in putting data from disparate sources onto the blockchain and running that data through linked applications. Such a platform will permit the creation of sophisticated integrated smart insurance contracts to cover complex risks and processes.
Insurers will replace multiple policies (with often-overlapping or gapped coverage) with a single, closed-form risk mitigation and claim adjudication solution.
For ease of reference, this paper will refer to these agreements between insurer and insured as Highly Integrated Smart Contracts (HISCs).
HISCs’ holistic approach to risk identification and mitigation allows insurers to become integral partners to their clients by offering a comprehensive solution that extends beyond risk transfer. Prevention of loss events will reduce costs and eliminate distractions, freeing insurers’ clients to focus on their core businesses. The HISC platform will also allow insurers to help clients
optimize their operations, resulting in increased efficiency, improved margins and greater operational certainty.
Let’s look at an example to see how this might work in practice.
Use Case: Semiconductors
Our use case will look at an application where sensors are already extensively used and tolerances are very tight: semiconductor fabrication.
Figure 2 is a high-level illustration of the HISC platform components shared by an insurer and a foundry, and some key points of interface and data exchange.
Sensor data related to the manufacturing and testing process as well as dynamic data on the physical and IT infrastructure of the facility are collected and sent to the blockchain.
Static data elements that factor into evaluating risk and potential liability – including licensing requirements, certifications and equipment specifications — are also sent to the blockchain.
The dynamic and static data are used in conjunction with the foundry’s internal limits, terms and specifications contracted with customers and regulatory requirements to form the rules of the HISC.
The engine evaluates all the data against the HISC rules and sends alerts when limits are approaching or breached, and similarly confirms when inputs and outputs conform to rules of the HISC. The AI layer of the engine would learn over time how the inputs affect one another in ways not previously recognized so that tolerances might be tweaked and interventions can occur ever earlier, further minimizing both risk and waste.
Platform in action
Let’s look at two scenarios to see how the HISC platform would work in practice. The first will look at risk mitigation, the second at claim adjudication.
Example 1: Risk mitigation
In a process as complex as semiconductor fabrication, there may be thousands of data points, but we will look at just one to get a sense as to how the platform might be used.
For the semiconductor to function properly, suppose each wafer must have a thickness of 525 millimeters with a tolerance of +/- 20 micrometers. However for one of the foundry’s customers, the manufacturer of implantable medical devices, a wafer more than 535 millimeters causes excess heat buildup and interferes with other critical components. The foundry has agreed to a tighter tolerance: +10/-20 micrometers.
Figure 3, illustrates the process for reporting the data on wafer thickness, evaluating the data against the rules and ordering and logging required actions.
Sensor data on thickness is sent to the blockchain and evaluated by the engine
The engine confirms that Chips 1-4 are within tolerance and sends that confirmation to the ledger. The engine determines Chip 5 is approaching the tolerance limit. (At this point, in optimized systems, relevant equipment would be re-calibrated before an out-of-tolerance wafer is even created.) Chip 6 is flagged as out of tolerance and is ordered destroyed. If Chip 6 completed the fabrication process and was shipped to the customer, a defective product may result. (The terms of the HISC may be such that product liability and other claims resulting from shipping that chip would not be covered.)
Chip 6 is destroyed, and a record of its destruction is sent to the blockchain.
Example 2: Claim adjudication
Arsine is an extremely flammable, explosive and toxic chemical used in the semiconductor fabrication process. After bringing a new machine into the plant with his forklift, Bob accidentally runs his forklift into an arsine storage vessel. There is a small fire that knocks the foundry offline for a week, and three employees are injured, including Bob.
Figure 4 illustrates the claim process through the HISC platform.
The foundry submits the claims to the insurer through the platform.
The claim triggers a request for the data related to the coverage requirements.
The engine evaluates the claim against the coverage requirements. It looks at Bob’s licensing and certification, maintenance records of the forklift, the specs of the storage container, the facility temperature and the performance of the fire-suppression equipment. The engine determines the claim is payable.
The insurer pays the foundry’s claim, a record of which is sent to the ledger.
These two hypothetical examples provide a sense of how these technologies will benefit insurers. Exactly how these benefits are realized will vary, but the benefit themes will be the prevention of loss events, reduction of operating costs, increase in efficiency and evolution of the insurer to an integrated services provider.
Below is an overview of the primary benefits we can anticipate from the HISC platform.
Fewer loss events
The HISC platform is designed with the express purpose of reducing loss events, which of course increases margins. The impact of loss events often extends far beyond the cost of the claim, so there is even greater value in avoiding them than is immediately apparent.
Lower operating costs for insurers
As even complex claims are auto-adjudicated, many of the costs associated with the claim process – investigation, audit, legal, sign-offs and disbursement – can be eliminated, significantly increasing margins.
Lower costs and greater efficiencies for clients
An optimized platform will let stakeholders know a problem is coming before it arrives. Alerts like approaching tolerance thresholds allows calibrations to be made before a breach, reducing waste and increasing margins.
Over time, the AI layer will provide new insights, enabling the engine to identify previously unknown correlations between inputs. For example, the engine may be determine that, if two particular specs both reach 90% of tolerance, failure is likely for a certain application. This insight will decrease returns and associated costs and increase customer satisfaction.
Increased customer satisfaction and higher retention
A less-than-smooth claim experience leads customers to move their business. HISCs provide friction-free claim payment, increasing customer loyalty.
HISCs allow insurers to create customized coverage, fully capturing the insurance expense, while clients are protected from coverage gaps without paying for products or features they don’t need.
New revenue opportunities for insurers
As insurers move beyond risk transfer and into other services, there will be new revenue opportunities for insurers. There are three broad opportunities in ascending order of complexity:
The massive amounts of data and insights from companies across their books can be sold to customers and service providers in benchmarking reports and industry studies.
Insurers will be well-positioned to provide operations and business process consulting services to their insurance clients.
Over time, small to mid-sized companies in certain industries may find it attractive to outsource manufacturing logistics management to insurers, akin to UPS’s move into supply chain logistics and management and Amazon’s into retail order fulfillment and cloud services.
Each of these additional revenue opportunities has the added benefit of increasing the stickiness of the customer relationship.
There are several non-inconsequential hurdles to be overcome before HISCs transform commercial and specialty insurance. Below are just five of the critical success factors necessary to implement HISC platform. Included is high-level estimate of the relative technical and implementation challenges of each.
1. Additional IoT data
Some operational and business processes may have dozens of data points to analyze, others thousands. The closer to 100% of the data that is brought on to the blockchain, the more effective HISCs will be. IoT data has grown significantly, but total capture rates vary widely. As sensor costs continue to fall rates will continue to increase. (Technical challenge: low to medium; implementation challenge: low)
2. Interoperability between different IoT platforms
Like any new technology, adding IoT to operational and business processes will be an iterative process. Within a single company, there are likely to be several different platforms for different processes (e.g. manufacturing processes, data security, equipment monitoring). IoT data from all of these different systems will need to be brought onto the blockchain and incorporated in to rules engines and other applications. (Technical challenge: medium; implementation challenge: medium)
3. Implementation of DLT standards
Neither HISCs nor the platform can be deployed on a public blockchain, so there must be some DLT standard broadly adopted by the industry. Corda (R3) and Hyperledger (Linux Foundation) are two consortium-led DLT platforms in development. JPMorgan (Quorum, built on Ethereum) and Monax have also created smart contract platforms. Companies are unlikely to support multiple standards, so getting agreement around one (or building a solution that is DLT-agnostic) is imperative. (Technical challenge: high; implementation challenge: very high)
4. Intelligent engine development
Putting data on the blockchain is simply readying the inputs. The data must be routed to a (likely) off-chain rules engines that must do more than provide a simple binary outcome. These engines will need to trigger actions, in some cases requiring human intervention, but still reported on the ledger. AI technology currently in broad use today is sufficient for the activities required in HISCs. The data integration, using ETL, data visualization, EDWs, etc., will be complex, but can all be accomplished using existing tools. (Technical challenge: low to medium; implementation challenge: low)
5. Integration of off-chain processes
Not every required action determined by the engine will necessarily be able to be executed on the platform. This may include backup communications between stakeholders, transferring funds and certain operational actions. These off-chain actions and their results will need to be cleanly brought back on-chain. Early on, there will likely be misses on some handoffs, but those challenges will be more due to planning and communication issues than technical shortcomings. (Technical challenge: low; implementation challenge: low)
The HISC platform described in this paper is closer to the end state than it is the next step of how these disruptive technologies will affect P&C insurance. The takeaway is that these (and surely other) technologies are an opportunity for insurers to not only change how they do business, but also the very nature of their business. They can become more than risk transfer counter-parties and create new revenue streams by offering high-value services, making them indispensable partners to their clients.
Getting there will not be easy. It will necessitate insurers becoming embedded, to a certain extent, with their clients. Successfully deploying HISCs will require insurers to expand beyond current competencies and develop expertise in other domains. These will include technology platforms, operations and logistics generally, as well as more specific subject matter expertise in the industries they serve.
This need to provide more integrated services to their clients may lead to strategic partnerships between insurers and consulting firms, or perhaps acquisitions of one by the other. Exactly how and how soon this unfolds is anyone’s guess, but, once change begins in earnest, momentum will build quickly, and insurers that are ill-prepared will find themselves unable to compete.
First, a definition. Distributed ledger/blockchain technology, increasingly abbreviated as “DLT,” transfers value in a decentralized, consensus-based and immutable manner using cryptographic tools and is different from technology today because it offers transactions occurring between unknown counterparties that are mathematically trusted in real time. DLT is at once a network and a database that can host applications like Smart Contracts, with the potential to be interoperable across trade ecosystems. This technology seems tailor-made to help administer the claims end of insurance.
Let’s talk about claims. It is well known that insurance claims are the storefront of an insurance business. Claims processing and resolution provide touchpoints for extended customer engagement, and a bad experience can poison an insurer in a customer’s mind, which can affect policy renewal. The claims experience should be seamless and easy to manage for all.
Imagine if you could smooth out your claims process so that it is more accurate, frictionless and cost-efficient and can even provide easy access to data for benchmarking and analysis to improve your customer’s digital experience.
I had my “aha” moment when I first learned about DLT technology. I was struck with an immediate vision of how things could be made better within the insurance industry. As a prior general counsel of an insurer, and now a consultant specializing in the strategic use of this technology, I understand how it can be implemented (once fully developed) and can envision how it can change and improve business from end to end.
Practically speaking, on the claims side, at the very least, the industry would never again have to suffer “the dog ate my homework” excuse for lost documents, duplicate or other document mishaps and related lawsuits. Claims provenance could be automatically established and adjudicated by so-called “smart contracts” (in the most general sense, they are protocols that have deterministic outcomes) in real time with an easily auditable and immutable trail. Identity proof would be less onerous. Those developments alone go a long way to reducing fraud and risk and their associated costs.
While modernizing claims processes is not a “sexy” thought, it is one that directly affects all insurers and their bottom lines by reducing risk. A small shift in the actuarial calculation based on a risk reduction goes a long way. There is not a business person on earth who does not want to increase revenue.
While there is a lot of hype, I believe we are only seeing the beginning of its potential. Education is needed. Imagination is needed. And innovation and execution are needed. The financial services industry has looked at this technology over the past year and is engaging with it, and some practical applications are expected to go into production in 2017. Insurers/asset managers should take notice. For instance, Delaware will begin using blockchain technology for UCC filings powered by Symbiont. Financial industry regulators, both domestically and internationally, are evaluating this technology and are listening and learning. In part, we owe the financial services sector a debt of gratitude for creating awareness overall.
Generally speaking, insurers have been slow to the table to learn about this technology, but it is imperative that they engage as early as possible because DLT has the potential to be very valuable for them. Some reinsurers already understand this and are experimenting. The diamond industry understands this and is experimenting with digital representation of hard assets on a blockchain for asset management and insurance purposes through Everledger. Other insurers have made some attempts to test similar concepts.
Indeed, the insurance industry can benefit on more than just the claims side.
We all know customer acquisition is the most uncertain and expensive part of the process in any business. Well-designed digital processes can prove invaluable in customer acquisition and retention. On the front end of the insurance industry, smart contracts can aid in creating easy-to-manage customer policies, which can be fed into databases and tailored and segmented in any way that makes business sense. Data management and security can be enhanced using blockchain technology. In fact, the Estonian company Guardtime has embraced the cyber security end of this technology and evolved a keyless signature infrastructure (KSI) that DARPA is verifying.
Blockchain/DLT technology is not a panacea for all. But it is worth exploring as the technology evolves. We are at an inflection point in the development of this technology—a point in time where insurers and others can have a say in how it evolves.Once standards emerge and practical applications are in production, it may be too late.
Time to get on board, insurers, and weigh in! All you need do is participate to make sure your interests are heard and accounted for.
To the insurance industry, I ask you: How do you see this technology affecting insurance?
A bitcoin (lowercase b), as a currency, has several flaws that will continue to limit its ability to replace money, as we know it. There are millions of words published on the subject, so I’ll leave it to the reader to assess arguments on both sides. However, Bitcoin (upper case B) as a protocol for the transfer of value is an extremely important innovation that the insurance industry would be wise not to ignore.
This article looks at the issue from the point of view where the insurance industry meets the engineering profession; this combination could be where some of the most important and valuable new opportunities arise.
The Block Chain Protocol (BCP)
The Block Chain Protocol is a brilliant innovation that cannot be un-invented – it is here to stay, and it will appear in many forms long after it sheds its association with so-called crypto-currencies. Bitcoin was designed to solve an age-old problem: the possibility of spending multiple times a promissory note such as currency. In the case of virtual currency, the problem is especially acute because a currency created on a computer can be easily copied by a computer.
The BCP can be compared to a train leaving the station. When the train arrives, the door opens and everyone piles in. After a predetermined amount of time, the doors close. While the doors are closed – and only while the doors are closed – the people write contracts for each other to agree upon. When the doors open, everyone piles out, but the contracts stay. Soon after, the doors close forever.
After the doors close, absolutely no changes can be made, ever. Any changes must be renegotiated as part of a new “block” in a continuing “chain” of transactions. This prevents someone from printing “money”, i.e., issuing the same contract to many recipients.
Today, this function is performed by a legal system, brokers and intermediaries such as banks and credit agencies – it is easy to see how these institutions would be concerned that an upstart technology that is fully decentralized with no CEO or corporate structure could literally exterminate their brokerage fees. (While I used a mechanical analogy of door and trains, the BCP operates using time stamps and cryptography to manage identities, ownership, vetting, etc.)
The big deal with bitcoin as a currency is that the value of a contract can be cast in time. The “crypto-currency” simply represents that value outside of the block for that exchange inside of the block.
Many people, including the media, get hung up on the idea of currency because that is something that obviously concerns everyone in the age of impending financial doom. However, one must not be fooled by hype nor remain complacent and hope the bitcoin issue it will go away. The BPC is here to stay, and there are thousands of them in existence, not just bitcoin.
Yes, this means threats to the status quo, but there are also great opportunities for those who learn how to use smart contracts to transmit value without institutional friction. The part that the insurance industry should be concerned with is the ability to transmit contracts.
When contracts are executed on a block chain and locked cryptographically, these are called “smart contracts.” The seminal work on smart contracts was written by Nick Szabo and introduced in this 1997 primer: The Idea of Smart Contracts. The remainder of this article will focus on one very important type of smart contract: the adjudicated smart contract partnering the insurance industry and the engineering profession.
Adjudicated contracts are contracts involving three parties: the insurer, the insured and the adjudicator. The insurance adjuster should immediately come to mind, but the work of the adjudicator is much more flexible.
In an insurance claim, there is often a forensic investigation involved. In many cases, the investigation may reveal failures of design, quality, defects and workmanship and moral hazard. When a payout is warranted, claim money is drawn for reconstruction and remediation per a contract.
The insurance industry depends on actuarial statistics and forensics to manage these risks. What if forensics could be performed and actuarial data compiled before the failure occurs?
Adjudication can be integrated directly into the performance contracts as the project is designed and built. Licensed professional engineers can “flip the switch” that releases funding or seal coverage for specific perils as they oversee the design/build contracts during design, construction and service life of a property. This would allow insurance companies the ability to price risk and adjust exposure pools with extreme accuracy.
Assurance by Design
In other words, it is possible to develop Block Chain Smart Contracts. My firm is doing this for the engineering, construction and property management industries. The concept is to codify current standard contract templates, such as AIA contracts, into a series of smart contracts on a cryptographic block chain. Contractual events will correspond to payment milestones underwritten by bank and insurance institutions. As each milestone is reached, the professional engineer will verify the proof of work and flip the switch that released the contract to the next insurable component.
The Insurance Industry Is Threatened
Today, many insurance companies are not too concerned with construction risk as long as it is priced correctly. What the insurance industry may not realize is that if too many good properties are subsidizing too many bad properties, private parties with good properties will use these adjudicated contracts to self-insure. For example, if a 250-unit, high-rise condo spends $4 million on a new potable water system and the insurance premiums are not discounted accordingly, the condo could now easily form its own risk-sharing pool with communities known to have new water systems.
With Block Chain Protocol technology and readily available data, almost anyone can now form an insurance pool.
The challenge then for the insurance industry is to use new technologies to build more and better insurance products using the legacy tools that they are built on and rapidly adopting new technological solutions that are now available to them.
The Block Chain Protocol may be one of the most important innovations of the digital age. Pretty much anyone with the job title of “broker” should be seriously concerned.
History provides countless examples of companies and industries that failed to adapt to new changes. For this reason, insurance should take the Block Chain Protocol very seriously. The technology is simple by design and only requires some creative adjustment and strategic partnerships to assimilate into the business plan.
Nobody will do it for us. We need to do it ourselves.