Examining The Benefits of The Institutes RiskStream Collaborative’s Proof of Insurance and First Notice of Loss Applications
Patrick Schmid, PhD***
Abstract
Blockchain and distributed ledger technology have potential to redefine insurance operations and help the insurance industry overcome many of its current challenges. The decentralized consensus process associated with blockchain or distributed ledger technology, what some call the “Trust Machine”, removes the need for intermediary verification and could dramatically lower costs. Competitors within the industry can join a network, like The Institutes RiskStream Collaborative , and securely share data with one another on a distributed ledger framework, like RiskStream’s Canopy Framework, abating duplicative efforts, minimizing reconciliation issues and reducing costs. The ability to use smart contracts — programmable code that can be written into a blockchain and self-execute — extends potential applications and makes automating large chunks of insurance-related processes more practical. This study analyzes the effects of instituting two of The Institutes RiskStream Collaborative’s personal lines auto insurance applications (Proof of Insurance and First Notice of Loss) within the U.S. market. The findings suggests that RiskStream membership could save between $19 million- $68 million in year 1, between $60 million-$190 million in year 2 and $99 million-$277 million in year 3 by instituting the applications. When production-grade blockchain or DLT use cases, like these, proliferate across the industry, the first entrants will be best positioned to understand the associated efficiency gains and reap the competitive rewards. Therefore, industry participants are best served engaging early in blockchain efforts, like the RiskStream Collaborative, rather than waiting on the sidelines.
LEARN MORE ABOUT THE INSTITUTES RISKSTREAM COLLABORATIVE HERE.
Table of Contents
Chapter 1. Introduction
Chapter 2. Blockchain and Distributed Ledger Technology
Chapter 3. The Institutes RiskStream Collaborative and the Canopy Framework
Chapter 4. Use Cases for Blockchain and Distributed Ledger Technology in Insurance
Chapter 5. Overview and Benefits of Application 1: Proof of Insurance (Personal Lines Auto)
6. Overview and Benefits of Application 2: First Notice of Loss (Personal Lines Auto)
Chapter 8. Potential Improvement in Customer Satisfaction from Application 1 & 2
Chapter 9. Total Benefits for Proof of Insurance and First Notice of Loss Including Customer Satisfaction
Chapter 10. Conclusions
Chapter 1. Introduction
The risk management and insurance industry is changing. The industry is faced with an increasingly fast moving, innovative, and data-driven environment, which may result in large-scale changes to traditional industry products, processes, distribution, and employment. New risks are emerging. New forms of data and analytics are changing the way the industry operates and analyzes risk. New tools are being utilized to create innovative efficiencies. The industry is embracing a variety of new technologies, including blockchain and broader distributed ledger technology, all at once.
Meanwhile, today’s economic climate presents many challenges for insurance-related organizations. In an extended period of weak income growth, rising prices, greater access to information, ever-evolving technology and increasing globalization, consumers demand more from suppliers, including insurers. Yet, in this increasingly competitive environment, profits have been constrained by low interest rates, weak investment returns and regulatory scrutiny. Insurance-related organizations have increasingly begun focusing on cost minimization in order to drive profitability. Much of this focus is on leveraging technology to lower the costs of recordkeeping, easing data retrieval, simplifying processes, combating fraud and finding an efficient path within a stringent regulatory environment.
As all of these factors play out and technological change flourishes within the risk management and insurance industry, blockchain and distributed ledger are playing a pivotal role. Most new technologies, including internet of things or machine learning, provide a means to capture and analyze data. Blockchain technology provides something different: a secure and permissioned way for entities to store and share data without the need for an intermediary. Up until recently, competitors within the insurance industry have had a fear of sharing data because the benefits did not outweigh the security risks and costs. That changes with the birth of blockchain technology.
Across the broad financial services industry, many are looking at blockchain technology and associated smart contracts as a potential opportunity to streamline the flow and verification of data, lower operating costs, improve processes and cut out the need for intermediation. Blockchain and distributed ledger technology may provide the industry with:
· Trust and auditability
· Increased automation
· Privacy with permissioned sharing
· Lower administrative costs
· Elimination of fraud
The potential use of this technology is one of the reasons the World Economic Forums stated that blockchain will soon become “the beating heart of the global financial system” and predicted that within ten years “10 percent of all gross domestic product will be stored on blockchains”.¹ Other groups have made similar predictions specifying insurance benefits. A McKinsey & Company report found that the insurance industry accounts for the most blockchain uses (22 percent of the total), distantly followed by the payments industry (13 percent).² A Deloitte report stated that “adopting a common blockchain across the sector could create a step-change in value in the insurance industry: claims handling could become more efficient and streamlined, resulting in an improved customer experience. Such an approach could also help to reduce further, if not entirely prevent, fraud if identity management was also enforced on the blockchain — meaning that criminals could no longer crash for cash.”³ Capgemini research indicates that personal auto insurers could save $21 billion a year through lower costs, which can be realized through the application of blockchain-enabled smart contracts.⁴
In this report, The Institutes RiskStream Collaborative demonstrates two practical use cases for distributed ledger technology. These use cases are on a path to production within the RiskStream membership. Cohorts have been formalized to test the underlying Canopy “distributed ledger” framework. This report provides an overview of blockchain and distributed ledger technology, an explanation of the RiskStream Collaborative and the Canopy Framework, a high-level analysis on blockchain use cases in insurance and a focused ROI analysis on the personal lines auto proof of insurance use case and the personal lines auto first notice of loss use case.
Chapter 2. Blockchain and Distributed Ledger Technology (DLT)
The blockchain and distributed ledger technology are significant in that they remove the need for verification by a central authority. For example, through its underlying technology, bitcoin solved the double-spending problem, which stymied digital currencies before it. It also reinvented the concept of monetary networks by providing a true peer-to-peer payment system and eliminating the need for intermediary banks, including central banks, in confirming and verifying transactions.
What was truly unique about bitcoin, and the blockchain behind it, was it provided a decentralization of trust. Traditionally, trust has been established by a centralized party, institution or intermediary. These parties, whether they are companies or governments, have been very important to establishing or giving root to our contemporary society. Yet, recently people or consumers who utilize these traditional systems have also been less trusting of these centralized institutions. Many trusted organizations have often misused consumer data and information. It turns out these systems can be untrustworthy at times. Cryptocurrency, particularly bitcoin, was important because it demonstrated that something critical in our society — the creation and transmission of money — could emerge without an intermediary involved in verifying transactions and establishing trust. Even the government is uninvolved. The idea of peer-to-peer transactional exchange is indeed a revolutionary concept.
However, blockchain applications are much larger in scope than bitcoin and the associated transaction protocol. More recent public blockchains, like the blockchain associated with the Ethereum Virtutal Machine (EVM), have further extended the blockchain innovation by establishing the use of smart contracts — code that facilitate, verify, and enforce the performance of a contract and that can be self-executing and self-enforcing. Ethereum allows codable contracts to be built and inserted into its blockchain so that contracts are enforced and verified without middlemen. Original blockchains, like bitcoin’s or ethereum’s function as shared databases (or more properly, ledgers) that are both public, in that transactions can be viewed by users, and anonymous, because the associated cryptography hides the identities of parties to the transactions.
It is important to distinguish between the concept of blockchain and cryptocurrency. Bitcoin and ethereum are examples of cyptocurrencies with their own public blockchains. Public blockchains are defined as blockchains where anyone on the system can read or write to the platform. There are nearly 2000 cryptocurrencies, each with their own blockchain or distributed ledger. These public blockchains have great revolutionary prospects, but their current potential offerings are bounded. Each of these public blockchains are fairly slow, since all transactions/smart contracts are broadcasted to all parties in the system, and have difficulty scaling. It may be the case that technological advancement continues and these hurdles are overcome, but currently they are not ideal for business usage.
Towards Other Blockchains and Distributed Ledger Technology
Although blockchain was born within cryptocurrency and the associated public blockchains, there are alternatives. Despite the rise in media attention regarding cryptocurrency and public blockchains, most businesses are focusing on private/permissioned distributed ledgers or consortia blockchains rather than public blockchains. For example, a consortia blockchain creates the potential for a shared ledger that, if adopted, could transform and automate countless traditional processes.
A private/permissioned distributed ledger or consortia blockchain could also help with trust. Businesses need to trust one another in order to interact and trade information. In fact, all parties within an industry, including government and regulators, trade information with one another as it is. Therefore, trust is desperately needed and it is also paid for. Businesses and governments are either storing data themselves, paying for intermediaries to store data for them, or both. If the organization stores data themselves, there is a cost to that data storage and there are transaction costs associated with sharing that data on demand. If the organization pays intermediaries, there is a storage cost and transactional cost with sharing and retrieving data. If they are doing both, there are reconciliation costs involved.
But, what if profit-seeking competitors within an industry trusted a secure blockchain or distributed ledger system to deliver necessary information to parties it deemed appropriate on a permissioned basis? Costs could fall for all parties substantially. Universal problems that affect an industry could be worked on in a collaborative manner through permissioned data sharing. All of the sudden duplicative efforts associated with transactions and various intermediaries would start to be eliminated. Reconciliation issues could eventually disappear. Permissioned regulatory reporting could be done in real-time with less resource drain.
For this reason, many within the insurance industry are turning their attention toward blockchain consortia because blockchain and distributed ledger technology can lower operating costs, increase automation, streamline usage and verification of data, improve processes, and eliminate the need for intermediaries. A realization is afoot that the blockchain or distributed ledger-based sharing works best within a robust network, so consortia are a logical starting point for adoption.
Chapter 3. The Institutes RiskStream Collaborative and the Canopy Framework
At its heart, distributed ledger technology is network driven. While the technology provides a means to work on the industry’s universal problems (i.e. fraud, uninsured motorists or reconciliation issues) that plague the industry and add costs, a non-partisan arbiter is needed to bring the industry together to test, learn and implement the technology.
As the name suggests, The Institutes RiskStream Collaborative emerged out of The Institutes, which is a not-for-profit formed over 100 years ago out of The Wharton School. The Institutes provides many services to the property & casualty insurance industry, but it is best known for its educational and research offerings. The Institutes educates more than 100,000 insurance professionals annually and has a board of CEOs who collectively represent a substantial majority of domestic insurance premium volume and a sizable international presence. The point being is that blockchain requires a network and The Institutes already has an established network.
In 2016, The Institutes started researching blockchain technology and became founding members of the Enterprise Ethereum Alliance. In early 2017, under the direction of The Institutes Board of Trustees, The Institutes launched blockchain working groups in proof of insurance verification, first notice of loss data sharing, subrogation — net settlement and parametric insurance. By the summer of 2017, The Institutes created a 501c6 non-profit called The Institutes RiskStream Collaborative. The RiskStream Collaborative operates as a consortium that uses its network to develop industry-specific blockchain and distributed ledger applications for varied use cases.
RiskStream members (carriers, brokers or reinsurers) are involved in leading all areas of RiskStream’s governance and activity. For example, members work with RiskStream staff to design use cases on behalf of the industry and then work with RiskStream staff and service providers to build out the use cases. RBA has program of solution providers consisting of organizations like Accenture, Deloitte, EY and Capgemini. These solution providers help by building out the RiskStream framework (Canopy), the use cases and ensuring the underlying distributed ledger is completely secure.
The Canopy Framework
Before Canopy, many insurance companies were creating their own blockchain applications, each with their own unique framework or personalized tweaks to various platforms. In a sense, each use case was operating in a silo. The industry wasn’t capitalizing on network effects or scale with this approach. Canopy was created to provide a backbone for various blockchain or DLT apps to be built upon.
RiskStream started by building proof of concepts on both public and private Ethereum — an open-source, blockchain-based distributed computing platform and operating system featuring smart contract functionality. The early work opened the RiskStream Collaborative’s eyes to blockchain’s potential to bring efficiencies to the industry. Nevertheless by late 2017, members voiced discomfort with storing hashed or encrypted policy or claims data in a fully distributed manner, despite the anonymity. There were concerns with metadata being used for competitive purposes. Instead, members voiced a preference for looking for alternative platforms that would allow member companies to share information only with other parties involved in that specific transaction.
In 2018, the RiskStream member-led Technical Committee went through an intense process of evaluating existing blockchain and distributed ledger platforms and selected R3’s Corda platform to allow RiskStream members to exchange data. The Canopy framework is built around Corda, but with a goal of being platform agnostic. It will be possible, long term, to use multiple blockchain or distributed ledger platforms with Canopy.
Chapter 4. Use Cases for Blockchain and DLT in Insurance
In early 2016, The Institutes drafted a paper on the potential applications for blockchain technology.⁵ This revealed that blockchain could have widespread ramifications across the insurance value chain, increasing market reach and customer personalization while also cutting costs. The industry could change in these ways:
• Insurance products, pricing, and distribution may be wildly altered as blockchain proliferation and its associated smart contracts spawn new products, such as parametric insurance and insurance implanted in transactional purchases, and realize efficiencies in the insurance process, thereby lowering prices and allowing for broader reach into emerging markets.
• Underwriting and risk management may see data-sharing capabilities and risk registries emerge. The immutability associated with blockchain provides provenance and auditability features. Peer-to-peer insurance models may also become more practical.
• Policyholder acquisition and servicing could become more efficient because new customer data will be increasingly confirmed at the origin. In addition, insurance life cycle documents will be easily updated with blockchain technology, avoiding repeat entry and verification and easing concerns with know-your-customer and anti-money laundering regulations.
• Claims management itself could be simplified through smart contracts, while an industry-wide shared ledger could help with multilayer settlements and fraud inspection.
• Finance, payments, and accounting in insurance could undergo considerable changes, as well as new opportunities. A distributed ledger, like blockchain, could allow for lower-cost international payments, more efficiency in subrogation via smart contracts, and new forms of raising capital.
• Insurance regulation and compliance could be transformed, as regulators would be able to monitor all insurance variables in real time and potentially create an industry-wide proof of insurance ledger.
Table 1 summarizes several potential use cases in each bucket of the insurance value chain.
If implemented, these blockchain use cases can benefit both insureds and insurers. From an insured’s perspective, industry use of blockchain may enhance the customer experience, improve affordability, provide a means for greater product innovation and allow for faster entry into emerging markets. From an insurer’s perspective, use of blockchain may lower costs, ease data retrieval, simplify processes, offer new products, combat fraud and lower regulatory burdens.
With the major benefits identified, the challenge now becomes actually realizing them by building blockchain and distributed ledger-related applications. This is a tall order for insurance organizations, given the number of evident use cases for blockchain and distributed ledger technology in insurance. Nonetheless, the build needs to start somewhere. RiskStream members started with Personal Lines Auto Proof of Insurance and Personal Lines Auto First Notice of Loss use cases.
Chapter 5. Overview and Benefits of Application 1: Proof of Insurance (Personal Lines Auto) Verification
Proof of insurance is required in a number of circumstances and often leads to costs for insurers as they field calls, exchange information, verify coverage and provide record-keeping services. In the United States alone, approximately 26.4 million people⁶ are involved in an auto traffic stop annually, and police report over 6.3 million auto crashes in a given year⁷. Each of these instances, totaling 32.7 million occurrences, involves auto proof of insurance validation — and likely represents a small portion of total auto-related proof of insurance verifications, which also include multiple vehicle crashes, registration checks, etc.
Distributed ledger technology, like blockchain, can help ease this process on consumers, agents/brokers, carriers and other interested parties by providing a single source of truth and a permissioned means to transfer insurance information across various parties. If a company is involved in a consortia network, like The Institutes RiskStream Collaborative, a DLT-based application could help the companies involved in the consortium to share data, cutting down on paper-related costs or data storage and the costs incurred by complying with state-based insurance verification systems as many of these state based systems mandate submission of complex data feeds. Longer term, if adopted by a large network, the proof of insurance use case could cut down on uninsured motorists, which represent about 13% of drivers. The costs of these motorists are generally passed onto insureds through their own UM coverage.⁸ The proof of insurance application may also have impacts on the customer experience by streamlining activities and easing the data sharing for the consumer.
RBA’s Proof of Insurance Use Case and Application Overview
Video Overview
For auto motorists, an insurer-provided paper card is the most common way to prove insurance coverage. The insurer incurs costs beyond the administrative costs for paper-based cards. State-based insurance verification systems mandate submission of complex data, which is extremely costly and often lacks real-time information. After exploring the motorist and customer journeys, it was apparent that if Proof of Insurance data was standardized and exchanged through distributed ledger technology, it will transform the Proof of Insurance process (exchange of information between policyholders or presentation to a third party) by introducing a unique way to validate and securely provide proof of insurance electronically in real time. Additionally, the seamless and secure transfer of insurance information for state and regulatory needs, can simplify compliance with state insurance verification systems and help identify uninsured and underinsured motorists — which is a growing problem in certain states.
Proof of Insurance — User Journey
The User Journey for Proof of Insurance has two pieces. The first piece of the proof of insurance application is an exchange with law enforcement. The second piece is an exchange between two different policy owners. The journey begins with the following personas:
Amy is policy owner A with Insure Co.
Jack is policy owner B with Acme Insurance
Joe is a police officer in Amy’s town
Part 1: The Personal Auto Insured to Law Enforcement User Journey
This part of the user journey involves Amy, a policy owner with Insure Co., who has either been in an accident or been pulled over by Joe, a police officer. Amy has been asked by the police officer to provide proof of insurance (Step 0). In order to provide proof of insurance details, Amy generates an access key via her mobile device (which leverages Canopy) and presents it to the police officer (Step 1 and 2). Joe, the police officer, enters the access key on his dashboard computer, which is integrated with RiskStream Network to check if the insurance policy is valid (Step 3). Through this process, the police officer receives a validation message that the insurance is valid (Step 4).
Part 2: The Personal Auto Insured to Insured User Journey
This part involves the exchange of insurance information between two parties (Amy and Jack) involved in an accident or incident. There is a need for the two parties to exchange proof of insurance details (Step 0). In order to provide proof of insurance details, policy owner A — Amy and policy owner B — Jack, each generate a unique access key and a QR code through their carrier’s mobile app (Step 1). One of them then scans the other’s QR code to receive the other party’s proof of insurance details (Step 2) in real-time, providing up-to-date and verified insurance information. Through this process, Amy and Jack’s proof of insurance details are automatically linked between both parties and the details are stored in respective carrier’s nodes⁹ including producer node if applicable (Step 3). Both policy owners, Amy and Jack, receive notifications and can view the other party’s exchanged proof of insurance details (Step 4). Either of them can retrieve and view the exchanged insurance details at a later point in time using the POI History feature (Step 5).
Proof of Insurance Benefit Analysis
The Benefit Analysis within this report focuses only on the impacts of the Proof of Insurance for personal lines auto use case within the United States. In addition only the known measurable impacts, are analyzed. There are likely other factors, including unknown measurable impacts, known unmeasurable impacts and unknown unmeasurable impacts associated with the use case.
A complete benefit analysis with involvement in the RiskStream Collaborative is not quantified in this report. Additional benefits would include — education, networking, marketing, involvement in governance, design thinking, involvement in working groups, framework usage and usage of all other applications.
Proof of Insurance: US Personal Lines — Automotive — Insured Estimates
The NAIC’s Auto Injury Database Report¹⁰, released in 2018, reports liability written exposures in the US for 2012, 2013, 2014 and 2015. U.S. estimates for 2016, 2017 and 2018, where created by averaging the annual growth for each area and forecasting that growth rate out. Chart 1 showcases an estimated insured motorists over time.
Chart 1
The count shown above is roughly in line with reports on the count of licensed drivers in the US, which according to the Federal Highway Administration stood at 221,711, 918 in 2016. In this report, we assume the count for 2018 of insured motorists is 225,670,179, as highlighted in Chart 1.
Proof of Insurance: Cost Reductions Associated with the Need to Store Image Files for Use Within Existing Mobile Apps
Most motorists in the US are required by law to maintain liability insurance coverage that will compensate others for injuries or damages resulting from auto accidents where the insured was at fault. All states, with the exception of New Hampshire¹¹, have adopted compulsory insurance laws requiring drivers to maintain liability insurance coverage. In most cases, proof of insurance is demonstrated via a paper card. Nonetheless, carriers have created mobile applications for policyholders and one feature associated with many of these applications is the ability to display proof of insurance. Forty-seven states allow verification via a mobile application.¹² The only states that do not allow for verification with an app are Connecticut, New Hampshire, New Mexico and District of Columbia. Despite the creation of these applications, many of the applications simply use scanned images of the paper cards. There is a cost to insurers to store these photographs within their systems. In Table 2, RiskStream attempts to create a rough estimate of the costs. An important note is we assume the images are stored on premises due the security aspects associated with the proprietary data including in an insurance card. This is important because the “fully-burdened” estimated costs of storage in the cloud are significantly lower than on premises. In fact, according to some research the fully-burdened cost of cloud storage could be one-fourth that of in-house storage.¹³
According to StorgeCraft (using a report by Gartner)¹⁴, the average cost of storing one TB of data in-house is $3,351. This number is supported in other research.¹⁵ It includes items like base cost, redundancy and backup, facilities and power, maintenance, migration and management.
Through an informal survey of RiskStream membership, it’s estimated that roughly 75% of policies have the ability to be shared with a mobile application, rather than a paper card. The constraints for this number are the states that do not allow it (mentioned above) and that certain insurance companies either don’t have an app or do not have a proof of insurance (POI) feature.
Table 2
From a rough sample of RiskStream members that did have mobile apps, only 20% did not use an image of the card to demonstrate proof of insurance via their carrier app. For the 80% that were using an image, this implies that the carrier is storing image files of every single policyholder whether they use the application or not. This, of course, comes with a cost.
Based on the assumption that liability written exposures serves as a reasonable proxy for policies, calculations are performed to obtain an estimated count of auto policyholders whose company’s app would store an image of their insurance information (estimated to be 135,401,107). If we assume that 160k image files are in one TB¹⁶, we end up with about 846 TB as the estimated total stored for these image files. Again, if the average cost of storing one TB is $3351 we simply multiply the two together to arrive at an estimated cost of $2,835,827. This is an industrywide cost that could be removed or reduced if an alternative method was considered, like DLT, and the entire industry adopted it. Of course, it’s unlikely that the entire network would adopt right away.
RiskStream produced an extremely conservative assumption on market share of application usage (network adoption) that will be used throughout this analysis. This expansion, which is showcased in Chart 2, is expected to be only a small part of the market (22%) in year 1 using the app in full production. However, there’s expected to be a significant increase in year 2 to 59%. By year 3, the network is expected to expand to 80% of market adoption. Again, these assumptions will be used in all models that follow.
Chart 2
If we apply the estimated network expansion to the potential annual savings of $2,835,827, we can obtain the storage cost savings for companies with apps in aggregate. We can use the network expansion to figure out what the expectations should be in aggregate by year. Chart 3 showcases these findings.
Chart 3
As noted above, the model dictates that the following cost savings could be witnessed just by reducing in-house fully burdened storage costs.
Table 3
Proof of Insurance: Reducing Costs of Providing Paper Insurance Cards
Auto insurance ID paper cards are issued with the policy, whether it’s a new or renewal policy. Since many auto policies are issued every six months, it’s possible that a ID card is shipped to each policyholder twice a year, and possibly more if changes are made to the auto insurance policy that trigger the issuance of new ID cards. Paper insurance cards can also be sent out by request. For example, if policyholders lose their insurance card, they may call to obtain a new one. An important piece of information not included in this particular analysis is quantifying the cost of an agent, broker, or carrier’s time associated with policyholders calling for a new card. This report does, however, try to quantify the cost of supplying the paper document.
This report estimates that the total cost of supplying a paper insurance card to a policyholder is about 2 cents, excluding any postage costs associated with those cards that are mailed separately based on policyholder request. Since multiple parties can be covered under one auto insurance policy requiring either multiple ID cards (or sheets of paper that they are printed on), this research utilizes a more conservative approach and assumes one printed card, or sheet of paper, per policy rather than multiplying the count by two since ID cards may be issued twice a year. So, similar to the last section on storage cost savings, this analysis begins with the total number of policies which were shown above in Chart 1. From there, some assumptions are made on the costs of supplying paper cards for proof of insurance.
Table 4
It’s assumed that the total cost of sending an insurance card is $0.02. If there are 225,670,179 insureds nationally and assume that the count of send outs is neither higher, nor lower than the insureds count, then one could simply multiply the two numbers to get the industry wide cost, which comes to $4,513,403. This implies if every company in the private passenger auto market were to do away with proof of insurance cards, there would be a $4.5 million savings.
However, insurers would not be able to immediately adopt this new option, even if they were all in the network. In Table 5 assumptions are made on the rate of expected “permission”. So, think of Table 5 as a table that assumes 100% of the RiskStream network could adopt, but only X% of the market were permitted to participate due to state law. This research makes very conservative assumptions on rate of participation by states, going from only 5% adoption in year 1, to 20% adoption in year 2, to 60% adoption in year 3.
Table 5
In addition to this, it should be kept in mind that the entire RBA network won’t engage right away. Therefore, an assumption of the network expansion is needed. Table 6 provides an update on the assumed network expansion and Chart 4 visualizes the expected benefit over time.
Table 6
Chart 4
Proof of Insurance: Reduced Costs of Complying with State Run Insurance Verification Systems
According to a reports by the Insurance Research Council¹⁷, various kinds of laws and regulations concerning auto insurance are expected to have an effect on the number of uninsured motorists in a state. For example, compulsory laws require all vehicle owners to purchase liability coverage, so they should reduce the number of vehicle owners driving without it. On the other hand, state laws that allow the stacking of UM limits might increase UM claim severity and frequency.¹⁸ State lawmakers have also introduced various penalties for uninsured motorists, including fines, suspended or revoked licenses or vehicle registrations, impounded vehicles, and jail sentences. The range of fines for uninsured driving vary. In many states, specific penalties for uninsured motorists are defined in the law, but are not mandatory or consistently enforced. In an attempt to identify uninsured motorists, some states have implemented insurance reporting requirements that leverage insurance information from carriers to spot uninsured motorists. This section of the report focuses in on the costs of these systems.
Complying with the various state run systems, each of which may have a different style or different data standards, present a significant burden and cost to insurance companies. It is important to note that when insurance data verification systems do not significantly reduce the number of uninsured drivers, insured consumers may also be negatively affected because the compliance costs to the insurers may be passed on to consumers. If so, insureds end up paying higher premiums to maintain these regulatory programs. Insured consumers also must spend time correcting state reporting errors, which depending on the system, are likely to transpire.
According to IICMVA research¹⁹ (cited below in italics), there are costs to various parties associated with existing systems:
The current reporting systems consume significant state and insurance company resources. Ongoing maintenance and operation of these programs require staff-intensive efforts by jurisdictions and insurers. Ultimately, these costs are borne by consumers.
Implementation Costs for State Jurisdictions
· The state of New York paid Anderson Consulting $4.5 million to implement its program. The project began in fiscal year 1999–2000.
· A 1997 audit conducted by the Utah Office of the Legislative Auditor General indicates the state spent $1.2 million to implement and administer its system when the reporting program was initiated in 1995.
· The Colorado Department of Regulatory Agencies (DORA) indicates the Colorado Motorist Insurance Identification Database (MIIDB) has cost the state approximately $7.1 million since 1997. The state employs eight full time equivalent (FTE) employees to manage the MIIDB program: one Office Manager and seven Administrative Assistant IIs. The state also pays a vendor to manage the database.
· The Missouri state reporting program is financed by an MIIDB Fund that collects 6% of the net General Revenue portion of the Insurance Premium Tax. As of June 2003, this Fund was collecting $3.2 million a year, but the Fund was not enough to cover the $3.7 million needed that year to maintain the system.
NOTE: The implementation costs identified above do not include revenues generated through fines by the state jurisdictions after implementation.
Costs for Insurers
· In 2000 it is estimated that the New York Insurance Information Enforcement System (IIES) cost four major carriers an average of $408,000 to develop and implement. There are approximately 300 insurance carriers in New York.
· Commercial automobile insurers spend $30 million annually to develop and maintain reporting programs.
· In one state alone, it has been estimated that commercial insurers spend $50 on database maintenance per insured vehicle. For example, a commercial fleet policy with 9,000 vehicles for a rental car company costs $450,000 to maintain the data reporting system each year.
· Negative publicity and customer experiences adversely affect policyholder retention.
· Considerable indirect expenses include legal, training, and public relations costs.
The cost to the industry is compounded by the fact that insurers are responsible for the development, implementation, maintenance, and administration of unique systems for each of the state programs.
Costs for Consumers
· Consumers may pay higher insurance premiums to offset insurer costs.
· Consumers as citizens pay for jurisdictional expenses via fees, assessments, and taxes.
· Insured drivers are fined inappropriately when mistakenly identified as uninsured.
There are several types of systems set-up for insurance verification. Each system has benefits and weaknesses. Chart 5 provides an overview of these systems.
Chart 5
A blockchain or DLT alternative could ease the burden on insurers to provide the information. The existing systems are:
1) Web Service Systems
There are 13 states with Web Service systems.²⁰ These systems are the most advanced and likely have the lowest cost to insurers to comply. However, there are still compliance costs with these systems. First, none of the systems are completely identical, so there are costs to provide information on a state by state basis. Second, pointers are still required to be sent out to the various states that leverage these systems. Third, in addition to the direct costs, there are also some indirect costs. For example, web service systems like these are much less secure then DLT or blockchain-enabled alternatives. Finally, these 13 state systems are not national and, therefore, can’t capture uninsured motorists from other states. The compliance cost of participating in a system like this would be ranked relatively low compared to other systems, like databases.
2) Database Systems
In this report, database systems are classified as those that require insurers to provide information about their books of business. These often require the insurer to submit an initial load of their book, followed by updates on a certain frequency (weekly, monthly, etc.). The idea is for states to compare insurer information to the state’s registration data to spot uninsured motorists.
The compliance cost of participating in a system like this would be ranked relatively highly.
3) Random Sampling/Other
Another effort used by several states is a random sample. The compliance cost of participating in a system like this is relatively low.
4) No System
Some states have no program. There would certainly be no compliance cost to insurers if there was no program in place.
Measuring the cost to complying with systems is challenging. The existing reports on the subject, mentioned in the IICMVA research above are roughly a decade or two decades old. Nevertheless, this research makes some conservative assumptions based other parties findings to attempt to quantify the potential costs.
First, it is assumed that costs will be functionally related to the amount of insured drivers within a state. In other words, the costs of compliance (whether using web services, databases or other) increase as the amount of insured drivers increases. Second, this research assumes that the costs per insured driver differ significantly.
· It is assumed that the cost of complying with a database program is 50 cents per insured driver. This is extremely conservative given previous research indicated it could be up to $50 per policy to comply (see IICMVA research above). Given the cited research is quite dated and was state-specific, it was assumed that the average across states was only 1/100th, or $0.5.
· It is assumed that states that fall into the web services alternative still have a cost, but much, much less at 0.1 cent per insured driver.
· It is assumed that the cost of complying with the random sample/other category is 0.5 cents.
· It is assumed that cost of complying with no system is zero.
[1] http://www.iicmva.com/IICMVAPublications.html
Chart 6 (*State by State Systems Need Confirmation)
This totals over $70 million across all states. Even if $70 million could be saved annually, not all states would allow an alternate version right away. Table 7 creates rough assumptions on state permission.
Table 7 below states adjusts the annual savings based on assumed rates of adoption. The assumed adoption rates by states are 40% in year 1, 60% in year 2 and 90% in year 3.
Table 7
The table also showcases the RBA network adoption rates that have been used throughout.
Table 8
Proof of Insurance: Reducing Costs Associated with Uninsured Motorists
According to recent data by the Insurance Research Council,²¹ approximately 13% of motorists in the US are uninsured. Motorists who forgo purchasing insurance create a problem that is of great concern to auto insurance policyholders, insurers, regulators, and the general public. In addition to paying for insurance that covers their own actions, insured drivers pay a portion of the costs incurred by drivers without insurance through uninsured motorist (UM) coverage. For insurers, costs associated with UM claims can be substantial.²²
An IRC report from 2014 attempted to quantify both the count of uninsured motorists and the costs.²³ According to the report, the 2012 count of uninsured motorists nationwide was 29,743,118. The report goes on to state that a true monetary estimate of the impact of these uninsured motorists to all groups is difficult, if not impossible, to achieve. In other words, cost estimate do not capture both direct and indirect costs to all parties involved. Nevertheless, certain UM-related cost estimates were produced and were used in the report to obtain quantifiable monetary estimates of the costs to insurers. For example, IRC’s Auto Injury Study²⁴ was used to quantify average payments for UM claims. In order to use the average payment, the report creates a methodology to estimate the amount of UM claims (included in the UM report’s Appendix). The report estimates there were 210,000 UM claims in 2012,²⁵ which translated to a national estimate of $2.6 billion in UM claims payments in that year.
Some Assumptions from the IRC report are shown in Table 9:
Table 9
If the same count of total motorists 236,056,492 is assumed as accurate (which is roughly in line with Chart 1), this count and the associated percentages (UM rate and insured rate) can be used to create new estimates if the UM rate were to change due to usage of the RiskStream Collaborative’s proof of insurance application. The question then becomes, how much could the proof of insurance application be expected to change the UM rate?
In this paper, it is assumed that the proof of insurance application would not impact the uninsured motorist rate more than no pay no play laws have been shown to. A recent study, published by the Journal of Insurance Regulation²⁶ and by the Insurance Research Council²⁷, documented potential effects of no pay no play laws on uninsured motorist rates. No pay no play laws prevent uninsured drivers from collecting compensation for noneconomic damages arising from a traffic accident with an insured driver. One objective of such laws is to minimize the incentives for driving uninsured, thereby decreasing the percentage of uninsured motorists. The report’s findings showed that states that instituted a no pay no play law did have a moderate (1.6 percent) decline in uninsured motorist rates the following year. It is not well known how the uninsured population found out about the laws in order to make the change.
This paper assumes that instituting the RiskStream Collaborative proof of insurance application would have similar impacts on the uninsured motorist rate as the institution of a No Pay No Play law (1.6%). This is a very conservative assumption as the No Pay No Play laws were only found to have only moderate impacts on uninsured motorists. RiskStream’s proof of insurance application could eventually be leveraged gain additional insights into the uninsured population and provide more of an active control mechanism. Therefore, the assumption of a 1.6% impact on uninsured motorists is likely too low, but does make sense as an estimation starting point.
As shown in Table 8, the estimation processes starts with the assumed number of total motorists published in the IRC report (236,056,492 motorists) and a new calculated UM rate. This calculated UM rate takes the original rate from the IRC UM study (12.6%) and adjusts it down by 1.6% (assumption from the No Pay, No Play Law study) to obtain 12.4%. These two numbers can be used to obtain a new count of uninsured motorists (29,267,228). Chapter 5 of the IRC’s 2014 UM report showcases the methodology to use other data (bodily injury claim frequency, number of UM causing bodily injury, etc.) and arrive at an estimated total for UM claims (209,004 UM claims below). The assumed average payment for a UM claim with bodily injury is taken from IRC’s Auto Injury Study ($12,381 per claim). In order to arrive at the new estimated total for UM claim payments, one simply multiplies the claim count (209,004) by the average payment ($12,381) to arrive at $2.587 billion. Please note, this is the estimated total UM payments “if” the UM rate declined by 1.6%, which we are assuming would occur with the institution of the RiskStream Proof of Insurance application similar to what has transpired with No Pay, No Play laws. The IRC’s UM report showed that the total UM payment was $2.6 billion under normal circumstances. Therefore, one can simply subtract the normal circumstance payment ($2.6 billion) from the new total payment ($2.587 billion) to arrive at the assumed savings associated with the Proof of Insurance application. The savings is $12.3 million, and again, this is a very conservative assumption.
Table 10
In addition to this, it should be kept in mind that the entire Riskstream network won’t engage right away. Therefore, an assumption of the network expansion is needed. Table 11 provides an update on the assumed network expansion.
Table 11
Totals for the Proof of Insurance Application:
This report has quantified four ways in which The Institutes RiskStream Collaborative’s Proof of Insurance Application for personal lines auto could lead to benefits. The analysis showed that it’s possible to have cost savings in reduction in data storage, paper cards, state by state verification system compliance and costs associated with uninsured motorists. Chart 7 and Table 12 demonstrate the total cost savings in each of these areas. The total potential savings are $9.5 million in year 1, $34.1 million in year 2 and $64.4 million in year 3.
Chart 7
Table 12
Chapter 6. Overview and Benefits of Application 2: First Notice of Loss (Personal Lines Auto) Data Sharing
According to recent auto claims statistics in the U.S., the number of auto bodily injury claims in a given year is 1.7 million, and the number of auto property damage claims is roughly 6.8 million, if you assumed they weren’t part of the same claim, the total is 8.5 million auto claims.²⁸ This number is much, much larger if aggregating claims across lines of business and totaling them globally. Regardless of the line of business, the first notice of loss experience doesn’t meet expectations for consumers: it should be more streamlined, personalized, seamless, and fast. For businesses (insurers, brokers, etc.), the current inefficient, manual process involves a large amount of iterative information exchange, wasted time/resources, irreconcilable recording keeping, and redundant completion of various forms.
A decentralized ledger provides the means to share data from insureds and insurers to the various involved parties (such as other insurers and collision centers) in a trusted manner without an intermediary while maintaining security through permissions. This can greatly improve the process, cutting costs for insurance-related businesses, which could be passed on to consumers. Perhaps, the greatest benefit, however, is the insurer to insurer exchange that can occur. According to RiskStream members, each insurer to insurer call related to a first notice of loss claims takes 15 minutes of time and there are several of these calls. With blockchain or distributed ledger technology, a single source of truth could be referenced, cutting down or eliminating these calls.
RBA’s First Notice of Loss (FNOL) Use Case and Application Overview
Video Overview
Today’s claim process drives excessive costs and customer dissatisfaction through long cycle times, inaccurate data, and haphazard information exchange between relevant parties. In a typical year, there are over 8 million auto claims in the United States, usually with multiple insurers, producers, and third-party service providers involved. Each one of these parties receives data at different times and different levels of detail related to the claim. This leads to significant back and forth between parties to collect, reconcile, and correct data, and ensure the claim is settled accurately. Significant opportunities exist to reduce both the true cost to process each claim by ensuring that more accurate data is available to more parties sooner in the process and overall loss costs by speeding claim cycle time and improving customer satisfaction.
Distributed ledger technology (DLT) will transform the first notice of loss (FNOL) process from a highly manual process coordinated by each insurer to a fully integrated process coordinated systemically by the RiskStream Network. After the insured reports the FNOL, this technology provides the means to share data from insurers and producers to the various involved parties (such as producers, other insurers, and collision centers) in a trusted manner while maintaining security through permissions. This can greatly improve the process, reducing costs for insurance-related businesses, which could be passed on to consumers.
[1] NAIC — Auto Injury Database
First Notice of Loss — User Journey
The User Journey for First Notice of Loss begins with the following personas:
Amy is a policy owner A with Insurer A
Diane is a policy owner B with Insurer B
Lauren is a claims adjuster for Insurer A
George is a claims adjuster for Insurer B
Jack is Diane’s broker for Insurer B
The Personal Lines Auto FNOL User Journey
This user journey involves a loss event between two or more different policy owners. In this case, an accident occurs between Amy and Diane (Step 0). If both Amy and Diane have the POI functionality on their carrier mobile app, they can efficiently exchange POI information via QR codes and the information is stored on the RiskStream Network. However, if both do not have this mobile app functionality, their POI information can also still be recorded and reported by traditional means of exchanging insurance cards (Step 1). In this case, Amy contacts her carrier, Insurer A, and Diane contacts Jack, her producer, directly via web, mobile app, or phone touchpoints to report the accident details (Step 2). FNOL data is collected from the policy owners and stored in the insurer and/or producer’s claim management system. This data is then sent to and stored in the RiskStream Network (Step 3). An attempt to identify duplicate records will also occur. If the other insurer(s) and producer(s) are known, the RiskStream Network shares the FNOL information to all involved stakeholders, writes the shared information to the other insurer and/or producer’s system, and provides notification. In this case, Insurer B would be notified (Step 4). The insurer and/or producer can then update any data elements in their own system, which is shared to the other stakeholders’ systems via the DLT, and a notification of these updates are also sent to all authorized parties (Step 5). Future functionality of this use case includes the ability of insurers, such as Insurer A and B, to query previous claims; identification, notification, and claim updates to third-parties, such as glass and auto body repair companies, and settlement approval (Steps 6–9).
First Notice of Loss (FNOL) Benefit Analysis
The Benefit Analysis within this report focuses only on the impacts of the First Notice of Loss for personal lines auto use case within the United States. In addition only the known measurable impacts, are analyzed. There are likely other factors, including unknown measurable impacts, known unmeasurable impacts and unknown unmeasurable impacts associated with the use case.
As mentioned earlier, a complete benefit analysis with involvement in the RiskStream Collaborative is not quantified in this report. Additional benefits would include — education, networking, marketing, involvement in governance, design thinking, involvement in working groups, framework usage and usage of all other applications.
First Notice of Loss: US Personal Lines — Automotive — US Claim Counts Estimates: Property Damage (PD) and Bodily Injury (BI)
The NAIC’s Auto Injury Database Report, released in 2018, reports claims for 2012, 2013 and 2014. This data excludes Texas. U.S. estimates for 2015, 2016 and 2017, including Texas²⁹, where created by averaging the annual growth for each area (BI and PD claims) and forecasting that growth rate out. Chart 8showcases the estimated claims in both bodily injury and property damage claims over time.
Chart 8
These are reasonable estimates for incurred claims, it’s possible that other claims are registered and this is a very low estimate for purposes of measuring first notice of loss. For this reason, we assume that the PD claims and BI claims are completely separate. So, in other words, this reports estimates total US claim counts in 2017 is 9.6 million (7.7 million PD claims plus 1.9 million BI claims). The report required a 2017 claim count, so that the report could leverage NAIC market share data later, which is 2017-based. The report assumes 2018 numbers would be similar to 2017 counts.
First Notice of Loss (FNOL): Assumptions on Time Spent Using Current Processes and Time That Would Be Saved with Implementation of RBA’s FNOL Application
Under the existing processes, insurer claim intake is quite lengthy. According to research conducted by Accenture, claim intake take about 15 minutes on average. The time spent sharing claim information or data from carrier to carrier also takes time. According to RiskStream membership, the time spent is approximately 12 minutes over average under current processes. See Chart 9.
Chart 9
Usage of RiskStream’s FNOL DLT application can lower the average time spent on these activities for various members by streamlining processes, eliminating manual data entry and reducing redundancy through a single version of the truth. According to Accenture’s analysis, time saved is estimated as 3 minutes on claim intake (a 20% improvement) and 6 minutes on data sharing (a 50% improvement). It’s likely both areas would improve further over time.
Chart 10
First Notice of Loss (FNOL): Generating Cost Savings
While it’s clear there are time savings, the cost savings is not yet approximated. In order to calculate cost savings, some assumptions are needed on the claim representative’s time. Leveraging data from Glassdoor, this report assumes an hourly salary of $30/hour (annual salary of $62,400 with benefits).
The average minutes saved on claim intake and data sharing are created into hourly estimates and multiplied by the estimated number of claims to arrive at the full potential hours saved with the DLT-enabled claims intake and data sharing processes. This process reveals that nearly 1.5 million hours could be saved annually in the US just on data sharing and claims intake alone.
Chart 11
If we assume a $30/hour rate, multiplied the potential for annual hours saved in both claims intake and data sharing, a total annual savings estimate is created. The estimate reveals a $14.4 million savings in claims intake is possible and a $28.8 million savings in data sharing is possible. This totals a $43 million in savings annually is possible.
Chart 12
First Notice of Loss: Conservative Assumptions on Market Adoption
The estimate above assumes, however, that all parties in the market are in the network and adopted the application. We make extremely conservative assumptions on market/network growth and adoption in the chart below similar to the assumptions made in the Proof of Insurance section earlier in the paper.
Chart 13
If we apply these numbers to the original “potential” annual savings, we can obtain estimated yearly savings. It is evident that estimated savings will grow as the RiskStream network grows and adopts.
Totals for the First Notice of Loss (FNOL) Application
This report has quantified two key ways in which The Institutes RiskStream Collaborative First Notice of Loss Application for personal lines auto could lead to benefits. The analysis showed that it’s possible to have cost savings in claims intake and data sharing. Chart 14 and Table 13 demonstrate the total cost savings in each of these areas. The total potential savings are $9.5 million in year 1, $25.5 million in year 2 and $34.6 million in year 3.
Chart 14
Table 14
Chapter 7. Total Benefits for Both Proof of Insurance and First Notice of Loss (Personal Lines — Auto) Applications in US
The analysis above provided estimated benefits of adopting personal lines auto applications for proof of insurance and first notice of loss. These are undoubtedly low estimations of all of the potential impacts, but do provide a starting point for further analysis. According to the results, The Institutes RiskStream Collaborative can expect a cost savings of $18.9 million for its members in year 1, which will expand to a cost savings of $59.6 million by year 2 and to $99 million by year 3.
Table 15
The analysis above did not include any elements of the improvement for customers. This section of the report attempts to quantify the potential improvement.
Chapter 8. Potential Improvement in Customer Satisfaction from Application 1 & 2
According to a new study by Pooser and Browne (2018)³⁰, customer satisfaction in auto insurance is linked to profitability. More precisely, J.D. Power data³¹ on customer satisfaction for insurers is correlated to lower operating expenses (likely lower expenses in retaining policyholders), which makes complete sense. The research shows, via regression analysis, that every 1 point improvement in the JD Powers customer satisfaction score is met with a corresponding decline of -0.136 percent in auto personal lines expense ratios.
There’s little doubt that a blockchain solution with both of RiskStream’s personal lines auto applications (POI and FNOL) could lead to a moderate improvement in the customer experience for several reasons. First, it could improve the experience of verifying coverage with law enforcement. Second, it may improve the experience of sharing insurance details with other parties when in an accident. Finally, the linkage that could occur through sharing these details, could expedite the claims intake process (covered in First Notice of Loss) and therefore lead to an improvement in customer satisfaction.
The Pooser and Browne research findings could be leveraged with private passenger data from the NAIC to quantify an improvement in the customer experience from utilizing the RBA applications. Several assumptions would need to be made to quantify potential impacts.
According to the NAIC, the private passenger auto industry net written premium is $222,234,919,000. Meanwhile, the expense ratio is 0.228. This indicates operating expenses of approximately $50,669,561,532.
The Pooser empirical analysis concludes that a one unit increase in the JD Powers score translates to a decline of -0.136 in the expense ratio. If we “assume a one unit increase in the JD Powers score could transpire from usage of both RBA applications, this could lower the expense ratio by this amount to 0.227 according to the study. The total operating expenses would then drop to $50,447,326,613. The difference between the original operating expense ($50,669,561,532) and lower operating expense ($50,447,326,613) is $222,234,919. Indicating, if the proof of insurance application and first notice of loss application could improve JD Powers score nationally by even one percentage point, it’s possible to save up to $222 million annually. A key assumption is that the customer service indicator nationwide would truly change by a full percentage point.
If it did change by this amount and there were rough assumptions on how quickly the RiskStream network would adopt, we could obtain some potential expense improvement estimates. Table 16 provides an overview of these expense improvement estimates for the better customer experience by year.
Table 16
Chapter 9. Total Benefits for Proof of Insurance and First Notice of Loss Including Customer Satisfaction
This report considers the customer satisfaction piece as upside potential for adoption of the two RiskStream applications. This information combined with Table 15 can yield the final ranges of potential market benefits from adopting RiskStream’s POI and FNOL applications.
Chart 15
Table 17
Chapter 10. Conclusions
It’s no surprise that insurance carriers, brokers and reinsurers are inspecting a variety of digital technologies, including blockchain and distributed ledger technology. A sizable portion of the insurance process involves transferring data. With distributed ledger technology, competitors within the industry can securely share data with one another on a permissioned basis, abating duplicative efforts, minimizing reconciliation issues and reducing costs. In addition, insurance-related organizations can leverage the shared ledger to avoid costly intermediaries and enact smart contracts in order to automate various procedures.
Blockchain technology is unique in that it is network-based. Although blockchain and distributed ledger technology could be leveraged within a single organization to bring various departments together, it is unlikely that operating a company-specific blockchain will be as productive as operating a blockchain that involves a larger network of competitors or participants. In order to get the most out of distributed ledger technology and leverage network effects, the industry must join together, working collaboratively and collectively to design holistic blockchain solutions. The Institutes RiskStream Collaborative is bringing brokers, insurers and reinsurers together to build distributed ledger solutions. While RiskStream has inspected a handful of use cases, RiskStream is currently focusing on bringing two personal lines auto use cases to production: proof of insurance verification and first notice of loss data sharing.
This report shared estimated benefits from early engagement in RiskStream’s Path to Production and adopting the developed solutions. These benefits far outweigh the costs, and therefore, are expected result in substantial return on investment for early adopters.
For example, in early 2019, RiskStream membership comprises well over 60% of private passenger auto insurance market share and that share is expected to grow. Yet, this report made conservative assumptions about membership adoption of RiskStream use cases, assuming only 22% of the market would be willing to adopt in year 1, only 59% would be willing to adopt in year 2 and only 80% would be being adopt by year 3. The report’s estimated savings for adopters is significant. In year 1, adopters are shown to save between $18.9–$67.6 million. In year 2, adopters are expected to save between $59.6–190.5 million. By year 3, these numbers were expected expand further with adopters saving $99.1–$276.8 million. Due to the potential operational improvements, industry participants are best served engaging early in RiskStream’s efforts, rather than waiting on the sidelines.
***Special thanks to RiskStream’s Products Team (Brendan Picha, Susan Kearney, and Sebastian Tilson) for their assistance in various aspects of this report. Thanks also to the Accenture team for their assistance with the First Notice of Loss data/model, the Deloitte team for their insight on Proof of Insurance, Alex Hageli of PCI for lending his expertise and providing information on insurance verification systems and the entire RiskStream membership for all of the help in creating this report. Much of the introductory sections of this report are leveraged from previously published publications: Blockchain Building Blocks (Schmid, 2016, updated Schmid, 2018), International Insurance Society Leaders of Tomorrow– Trust Link (Schmid 2018) and What Will Blockchain Mean for Insurance (Schmid 2018).
Citations/Footnotes
[1] Peter Vanham, “Blockchain Will Become ‘Beating Heart’ of the Global Financial System,” World Economic Forum, August 12, 2016, https://www.weforum.org/press/2016/08/blockchain-will-becomebeating-
heart-of-the-global-financial-system/ (accessed January 23, 2017).
[2] “Blockchain Technology in the Insurance Sector,” McKinsey & Company.
[3] “Blockchain Applications in Insurance,” Deloitte, 2016, https://www2.deloitte.com/content/dam/Deloitte/ ch/Documents/innovation/ch-en-innovation-deloitte-blockchain-app-in-insurance.pdf (accessed January 23, 2017).
[4] “Consumers Set to Save Up to Sixteen Billion Dollars on Banking and Insurance Fees Thanks to Blockchain-Based Smart Contracts Says Capgemini Report,” Capgemini, October 11, 2016, https://www.capgemini.com/news/consumers-set-to-save-up-to-sixteen-billion-dollars-on-banking-andinsurance-fees-thanks-to/ (accessed January 23, 2017).
[5] “Blockchain Building Blocks”, The Institutes/Schmid, https://www.theinstitutes.org/doc/riskblock/Blockchain_Building_Blocks.pdf (accessed January, 2017)
[6] Bureau of Justice — Traffic Stop Data
[7] NHSTA — Motor Vehicle Crash Statistics
[8] “Uninsured Motorists”, Insurance Research Council, 2017 https://www.insurance-research.org/sites/default/files/downloads/UMNR1005.pdf (accessed December, 2018)
[9] Nodes are participants systems connected to the network.
[10] NAIC’s Auto Injury Database Report, 2018 https://www.naic.org/Releases/2018_docs/naic_releases_auto_homeowners_reports.htm (accessed December 2018)
[11] In New Hampshire drivers who have demonstrated lack of financial responsibility for past auto accidents must prove proof of financial responsibility.
[12] https://www.policygenius.com/blog/which-states-let-you-use-an-app-as-proof-of-insurance/
[13] https://cdn.newswire.com/files/x/b7/6d/37393813af72d57abd94c569ab0e.pdf
[14] https://blog.storagecraft.com/file-storage-cost-statistics/
[15] https://cdn.newswire.com/files/x/b7/6d/37393813af72d57abd94c569ab0e.pdf
[16] https://www.quora.com/How-many-picture-files-can-fit-in-1-TB-of-storage
[17] Insurance Research Council, Uninsured Motorists, 2014 Edition (Malvern, Pennsylvania: The Institutes, 2014), http://www.insurance-research.org/research-publications/uninsured-motorists-2014-edition (accessed March 22, 2017).
[18] The objective of UM stacking is to provide greater reimbursement to the accident victim. Stacking provisions are typically designated as “intra-policy stacking” or “inter-policy stacking.” Intra-policy stacking involves combining the limits of the vehicles insured under the same policy, in which case the limit is multiplied by the number of vehicles insured. Inter-policy stacking involves combining the limits of vehicles insured under multiple policies.
[19] http://www.iicmva.com/Making%20V2.1.docx
[20] http://www.iicmva.com/IICMVAPublications.html
[21] IRC Uninsured Motorists Reports (cited earlier)
[22] In 2012, the average UM claim included $14,359 in economic losses due to auto injuries and an average total UM injury payment of $12,589, not including payments for property damage. Insurance Research Council, Auto Injury Insurance Claims: Countrywide Patterns in Treatment, Cost, and Compensation (Malvern, Pa.: Insurance Research Council, 2012) p. 39.
[23] IRC Uninsured Motorists Report (cited earlier)
[24] Insurance Research Council, Auto Injury Insurance Claims: Countrywide Patterns in Treatment, Cost, and Compensation (Malvern, Pa.: Insurance Research Council, 2012).
[25] This estimation included bodily injury claims only. It’s likely higher with property damage claims included.
[26] Schmid, Journal of Insurance Regulation; Winter2013, Vol. 32, p83 http://connection.ebscohost.com/c/articles/92960639/potential-effects-no-pay-no-play-laws
[27] https://www.insurance-research.org/research-publications/potential-effects-no-pay-no-play-laws
[28] NAIC Auto Injury Database Report, 2017 https://www.naic.org/prod_serv/AUT-PB-14.pdf
[29] This report assumes Texas share of claims is aligned with the state’s share of countrywide exposures (assumed to be 7.6%).
[30]St. John’s University, “Profitability Across the U.S. Insurance Industry,” Professor David Pooser and Professor & Faculty Chair Mark Browne, 2017–2018
[31] J.D. Power, 2016, U.S. Auto Claims Satisfaction Study. Available at http://www.jdpower.com/resource/jd-power-us-auto-claims-satisfaction-study (accessed February 19, 2017).
