Insurers face a myriad of challenges in today’s environment, routinely returning less than their cost of capital. Rising labor costs, increases in customer expectations, additional regulation of financial markets and pressure from digital native insurtechs combine to give even the most calm executive stress.

Given the nature of the high-volume and data-rich business, many would suppose insurers would quickly turn to automation and AI to combat some of their challenges. However, to date, insurers have been hampered by the inability of traditional automation technologies like robotic process automation (RPA) to handle the complexity and variability of the industry.

Thankfully, with the recent rise of natural language processing (NLP) models (including large language models, or LLMs) insurers can now deploy automation widely across their value chain to reduce back-office and front-office costs. The changes boost employee productivity and happiness by enabling them to focus on what humans (and not computers) do best and improve customer experience by allocating more employee time to customer engagement and providing greater personalization to meet customers' needs. 

See also: The Risks of AI and Machine Learning

Why Haven’t Insurers Automated More?

Automation often has the highest return on investment on processes that contain: high volumes, lots of manual effort, multiple applications, tight service-level agreements (SLAs) and severe penalties for errors. By this definition, the insurance industry should be the perfect fit for automation and AI adoption, yet struggles to adopt on a large scale. Why?

Certainly, risk aversion can make insurers conservative. Yet, this fails to explain why insurance companies aren’t turning more to AI and automation to solve their challenges as other conservative industries (manufacturing, logistics, etc.) are widely using AI, especially in the back office.

Instead, the reason the insurance industry has not seen an explosion of AI implementation stems from the lack of capability in traditional automation systems like RPA. RPA requires:

1. Highly Standardized Processes
2. Exceptions to the Rule Anticipated Up Front
3. Large Workforce of Special RPA Developers
4. Constant Communication Between Developers and Subject Matter Experts (Like Claims Agents).

Unfortunately, these requirements often prevented many processes from being suitable for automation, especially processes like claims. A typical claim may contain a litany of different process “paths” based on the unique situation. In process terms, these are called “exceptions.”

Traditional automation is unable to understand the complexity of such situations or handle all the various edge cases that a human can easily handle. Adding to this problem, traditional automation requires hiring armies of “RPA developers,” highly paid individuals able to take a proposed process and translate it into the programming language (even if drag and drop) that automation can understand. Exceptions contained in most insurance processes, including customer onboarding, claims processing and underwriting, mean developers would be constantly required to try and program each and every one of the edge cases. In the past, it made more sense to simply outsource many of the processes to offshore centers.

This left insurers in a bind. Labor costs (including outsourcing) are rising faster than premiums. In addition, by outsourcing these processes insurers are unable to profit off the rich data created when such processes are digitally recorded. Thankfully, new AI and automation has arrived.

See also: Insurers Boosting Their Use of AI

A New Era: The Subject Matter Experts in Control

The rise of natural language processing and LLMs like ChatGPT have eliminated one of the biggest headaches with AI and automation: the need for developers. Instead, new automation systems are built with NLP at the core – making the building and management of even highly variable, complex automations accessible for the subject matter experts. Doing so eliminates many of the problems insurance processes have typically faced.

1. Even exception-heavy processes like claims can be automated

Subject matter experts like call-center agents, brokers and underwriters know their business well. If something unexpected occurs, more times than not they’ve encountered it before. Because automation is now all in English, new technologies make it easy for the subject matter experts to handle any exception simply by conversing with AI. Now, even complex processes with lots of variability in documents or rules (like insurance claims) can be automated, driving significant cost savings through employee reassignment or capacity creation for growing segments.

2. The deep industry knowledge and experience of an aging workforce can be captured as a rich data source.

In today’s data-driven economy, knowledge of a business’s processes can yield significant efficiency gains in process re-design and improvement. However, because of the vast quantity of scenarios and incidents that can occur in the coverage of insurance businesses, capturing a view into the processes can be too expensive relative to the ROI. However, with NLP-based automation, the automated activities (and the SME’s responses to exceptions) are recorded in English in the form of a “business journal.” This data represents a rich treasure-trove of data on which prompt based queries can be run using LLMs. Executives can now gain on-demand answers to core questions about their business, notice trends and make more effective decisions. It’s precisely here where legacy insurance providers have an advantage relative to entry insurtech players. The volume of processes, if captured properly, can provide far more market data at scale than new players have, providing essentially key market research.

3. Front-office employees can spend more time focused on customer relationships.

Customer service scores continue to decrease in the insurance industry across sectors. This decline stems from frustration in claim creation, delays to claim resolution and a feeling of the customer “just being a number.” Now, NLP-based automation technologies solve these problems. In a recent example, one contact center saw a decrease in average handle time (AHT) from 513 seconds to 209 seconds by automating the account lookup, note taking (from voice to text) and primary actions taken. With this sort of automation, insurers can enable their people to genuinely focus on their customers, develop trust, provide personalized care and improve both employee and customer experience alike.

The new era of NLP based automation provides a rich opportunity for an industry struggling to contain costs, improve customer experience and fight off new competition. Traditional limitations to legacy automation systems like RPA are unlocked by NLP-based automation. Now, automation and AI are accessible to insurance subject matter experts, boosting productivity and allowing for even highly variable, complex processes to be automated.

As more is automated, variable costs will decrease significantly in this high-volume industry, data relative to the working and market will be captured efficiently and a better customer experience will drive retention, cross-sell and up-sell opportunities.

KEY TAKEAWAYS:

--Allianz Commercial analysis of large cyber losses shows the number of cases in which data is exfiltrated is increasing every year – doubling from 40% in 2019 to almost 80% in 2022, with 2023 significantly higher.

--The frequency of cyber claims will increase about 25% this year.

--Ransomware activity alone was up 50% year-on-year during the first half of 2023. So-called Ransomware-as-a-Service (RaaS) kits, where prices start from as little as $40, remain a key driver. Ransomware gangs are also carrying out more attacks faster, with the average number of days taken to execute one falling from around 60 days in 2019 to four.

----------

Following two years of high but stable losses, 2023 has seen a worrying resurgence in ransomware and extortion claims as the cyber threat landscape continues to evolve, Allianz Commercial warns in a new report.

Hackers are increasingly targeting IT and physical supply chains, launching mass cyber-attacks and finding new ways to extort money from companies, large and small. Most ransomware attacks now involve the theft of personal or sensitive commercial data for the purpose of extortion, increasing the cost and complexity of incidents, as well as bringing greater potential for reputational damage. Allianz Commercial analysis of large cyber losses shows the number of cases in which data is exfiltrated is increasing every year – doubling from 40% in 2019 to almost 80% in 2022, with 2023 significantly higher.

Cyber claims frequency has picked up again this year as ransomware groups continue to evolve their tactics. Based on claims activity during the first half of 2023, we expect to see around a 25% increase in the number of claims annually by year-end. The attackers are back, and focused again on Western economies, with more powerful tools, enhanced processes and attack mechanisms. Given this dynamic, a well-protected company is necessary to stand up to the threat, and, increasingly, the most important element of this is developing strong detection and fast response capabilities.

See also: Role of Ransomware in Cyber Insurance

How is ransomware risk evolving?

According to the Allianz Commercial report, "Cyber security trends 2023: The latest threats and risk mitigation best practice – before, during and after a hack," the frequency of cyber claims stabilized in 2022, reflecting improved cyber security and risk management actions among insured companies. Law enforcement agencies targeting gangs, together with the Ukraine-Russia conflict, also helped curtail ransomware.  However, ransomware activity alone was up 50% year-on-year during the first half of 2023. So-called Ransomware-as-a-Service (RaaS) kits, where prices start from as little as $40, remain a key driver in the frequency of attacks. Ransomware gangs are also carrying out more attacks faster, with the average number of days taken to execute one falling from around 60 days in 2019 to four.

Data exfiltration can significantly add to the cost of a loss or cyber claim. Such incidents can take longer to resolve, while legal and IT forensics can be extremely expensive. If data has been stolen, companies must know exactly what data has been exfiltrated and will likely have to notify customers, who could seek to claim compensation or threaten litigation.

This year has also seen several large mass ransomware attacks as threat actors used exploits in software and weaknesses in IT supply chains to target multiple companies. For example, the MOVEit mass cyber-attack, which exploited a data transfer software product, affecting millions of individuals and thousands of companies, contributed to the increase in the frequency of claims in 2023 to date, affecting multiple policyholders simultaneously.

Growing number of public cases

In the past, the number of cyber incidents that became public knowledge was low. Today, hackers threaten to publish stolen data online. Allianz Commercial analysis of large cyber losses (€1mn+) shows that the proportion of cases becoming public increased from around 60% in 2019 to 85% in 2022, with 2023 set to be even higher. 

With potentially costly financial and reputational consequences, companies may feel under more pressure to pay ransoms where data has been stolen. The number of companies paying a ransom has increased year-on-year – from just 10% in 2019 to 54% in 2022, again based on analysis of large losses only (€1mn+). Companies are two-and-a half times more likely to pay a ransom if data is exfiltrated, on top of the encryption.

However, paying a ransom for exfiltrated data does not necessarily resolve the issue. The company may still face third party litigation for the breach of data, especially in the U.S. Indeed, there are few cases where a company should believe that there is no other solution other than paying the ransom to be able to re-access its systems or data. Any impacted party should always inform and cooperate with the authorities.

See also: Tackling the Surge in Cyber Premiums

The importance of early detection and fast response

Protecting an organization against intrusion remains a cat-and-mouse game, in which cyber criminals have the advantage. Allianz analysis of more than 3,000 cyber claims over the past five years shows that external manipulation of systems is the cause of more than 80% of all incidents. Threat actors are now exploring ways to use artificial intelligence (AI) to automate and accelerate attacks, creating more effective AI-powered malware, phishing and voice simulation. Combined with the explosion in connected mobile devices – Allianz Commercial has seen a growing number of incidents caused by poor cyber security in this area – attack avenues only look likely to increase.

Preventing a cyber-attack is therefore becoming harder and the stakes higher. As a result, early detection and response capabilities and tools are becoming ever more important. Around 90% of incidents are contained early. However, if an attack is not stopped in the early stages the chances of preventing it becoming something much more serious and costly greatly reduce.

Companies should direct additional cyber security spending on detection and response, rather than just adding more layers to protection and prevention. Only one-third of companies discover a data breach through their own security teams. However, early detection technology is readily available and effective.  Cyber breaches that are not detected and contained early can be as much as 1,000 times more expensive than those that are, with Allianz Commercial analysis showing that early detection and response can stop a €20,000 loss turning into a €20mn one.

The buzz surrounding generative AI has permeated every industry, and the insurance sector is no exception. With capabilities such as creating human-like responses and generating insightful outputs, generative AI technologies have brought forth a unique blend of opportunities and challenges for organizations.

From streamlining various business operations, such as content creation, to spurring debates about its potential impact on employment, generative AI has firmly embedded itself in the strategic discourse. Paul Carroll, editor at Insurance Thought Leadership, emphasized in his May 2023 commentary, “Generative AI: Coming Faster Than You Think,” that a balanced perspective on the short-term and long-term implications of these technologies is vital for shaping future business paradigms in the insurance industry. He writes, “The rule of thumb about breakthrough technologies is that they’re overestimated in the short term but underestimated in the long term.”

Understanding generative AI necessitates a look beyond its surface-level capabilities. It is pivotal to comprehend that these models do not “think” autonomously; rather, their outputs mirror the quality of their training data and the effectiveness of human-generated prompts. Therefore, there is and will be a constant need for a human-machine loop to exist and work together.

See also: 5 Ways Generative AI Will Transform Claims

Harnessing the Power of Language Models in the Insurance Industry

Generative AI, particularly through large language models (LLMs), can generate coherent and seemingly intuitive responses to user queries. This is achieved by training on vast datasets, understanding grammatical structures and learning word sequences, which enable the AI to predict forthcoming words in a sentence and familiarize itself with industry-specific terminologies.

Generative AI introduces transformative capabilities to the insurance industry, offering innovative solutions across various verticals including customer service, risk management, product development, claim processing and marketing. It enables the creation of sophisticated, personalized customer experiences through intelligent communication. 

This enhanced responsiveness to human input is widely considered to be one of the most significant advantages of generative AI. Rather than reinventing the wheel each time a new communication or document is needed, these technologies can draw from an insurer’s entire library of archived communications in a matter of seconds, generating a strong initial draft that can then be refined by a human author.

Despite the myriad applications, it is imperative to navigate the implementation of generative AI with a keen emphasis on ethical use, regulatory adherence and data security.

Enhancing Customer Communications Through Sentiment and Readability Adjustments

To the general consumer, selecting the right insurance product can be confusing and stressful. Insurers must take great care with their communications to ensure they are easily understood. In times of great distress, even more care needs to be taken. When a denial of claim or coverage is required, using distinctly negative language, such as the words “no,” “illness,” “poverty” and “death,” have been shown to release stress hormones in the brain of the person reading it. These stress chemicals impair judgment and can reduce reading comprehension by up to four school grade levels, compounding the challenges faced by consumers when communications are not clear. Generative AI can step in to support the rewriting of content in plain language or to align to a targeted reading level, which is important given that half the U.S. population reads at or below an eighth-grade reading level.

Generative AI, by leveraging natural language processing (NLP) for sentiment analysis, can help adjust the emotional tone of communications, ensuring that even negative messages are conveyed in a manner that will be received as well as possible.

See also: 3 Key Uses for Generative AI

Amplifying Business Needs With AI: The Imperative of Choosing the Right Partner

Navigating through the intricate pathways of generative AI, particularly in the context of the highly regulated and nuanced insurance sector, brings forth several challenges that can be mitigated by aligning with a skilled technology partner. The deployment and management of generative AI entail not only a deep understanding of the technology itself but also an awareness of the regulatory, ethical and data security nuances associated with its application in specific use cases within the insurance domain. A proficient AI technology partner, with a wealth of expertise and experience across varied implementations, can deftly navigate through these complexities, ensuring the AI applications are compliant, secure and effectively tailored to meet specific organizational objectives.

Moreover, the insurance landscape is characterized by dynamic shifts influenced by regulatory changes, market trends and evolving customer expectations. Engaging with an adept AI partner ensures the generative AI applications are not only attuned to the current needs of the organization but are also scalable and adaptable to accommodate future evolution.

The partner can facilitate continuous learning and adaptation of the AI models, ensuring they evolve in tandem with shifting trends and regulatory norms, thereby securing a sustainable and forward-looking AI strategy. Implementing AI is not a one-off project, but a continuous journey of learning and adaptation. 

The World Economic Forum estimates that by 2025 approximately 463 exabytes of data will be produced every day. To put that number in context: All the words ever spoken in human history can fit into FIVE exabytes.

Data is powering disruptive business models that account for sizable portions of our modern economies.  

Data as we know it has gone through a number of transformations since the computer era began. Between 1969 and 2000, an era we'll call Data 1.0, data was largely confined to specific applications, such as online booking systems and payroll automation. The era of Data 2.0 arrived when business leaders realized that by sharing information across the organization, workers could be empowered to make more-informed decisions. This era sparked advancement in analytics and data warehousing and the rise of new methods, such as master data management (MDM).  ‍

The advent of social media and cloud computing and the explosion of mobile has now given rise to Data 3.0. With more data being produced than ever before, digital transformation is rapidly creating the businesses of the future.  

With 4.8 billion (yes, billion with a “b”) internet users, people are generating data every minute of every day. But it’s not just us as individuals who are creating all this new data. The rise of Internet of Things (IoT) devices, which are able to connect to the cloud and share data with other devices, will mean connected humans will be massively outnumbered by connected machines, with some predicting 75 billion connected devices by 2025.  

‍Disruptive businesses, such as Airbnb, Amazon and Uber, harness data and use it as the very architecture of the services provided. If traditional sectors, like insurance and banking, hope to keep up, it is imperative to find ways of leveraging the available data to deliver a better experience, a better product or even a whole new way of doing business. 

See also: Why Becoming Data-Driven Is Crucial

You Can’t Use It if You Can’t Find It (Share It, or Trust It)

The growing reliance on data has taken the conversation about access and management out of IT and placed it firmly in the boardroom. CEOs now realize data management is as critical to future success as finance and talent management. Most importantly, leaders now understand that effective decision-making is based on accurate data.  

Unfortunately, many organizations face a real challenge, with data trapped in separate systems and silos, both internally and off-site, and in a variety of forms. A company may be generating terabytes of useful information, but it can’t be analyzed if it can’t be located. Data fragmentation has become one of the biggest operational challenges facing companies looking to accelerate digital transformation.  

‍More than just preventing good internal decision-making, fragmented or poorly managed data can affect how a business interacts with partners and within ecosystems. In a recent survey, Gartner found that 65% of decisions made are now more complex than just two years ago and involve more stakeholders than ever before. Gartner says that decisions need to “...become more connected and that sharing of data and insights across organizational boundaries is critical.” 

In fact, Gartner predicts that in 2023, organizations promoting data sharing will outperform peers on most business value metrics. However, this compelling opportunity seems out of reach for most companies because, according to Gartner research, less than 5% of data-sharing programs can currently correctly identify trusted data and locate trusted data sources.  

See also: Data Mesh: What It Is and Why It Matters

Taking Back the Power 

When it comes to operationalizing the many benefits of Data 3.0, leaders need to first look at ways to regain control of existing or internal data. This will help optimize decision-making processes, as well as maximize the many benefits of working more closely with ecosystem partners. ‍

Findings of the Dresner Advisory Associates’ 2020 Data Pipelines Market Study published in Forbes show that over 80% of enterprise business operations leaders say data integration is critical to operations. What’s more, 65% of organizations prefer to deploy data integration solutions from cloud platforms or hybrid cloud.  

Modern hybrid integration platforms (HIPs) offer complex data processing with simplified usage and are simpler, faster and more powerful than traditional legacy-bound data solutions. HIPs can also be customized to meet specific goals, to reduce the need for businesses to invest in highly skilled developers and tech support teams by simplifying multiple data management functions and to operationalize data for other systems to use while continuing to advance the company’s broader digital transformation agenda.

There is no doubt that managing data is swiftly becoming one of the most important jobs in every business. There is already an extreme shortage of data scientists across industries, and the gap is growing every day.

With every human producing an average of 1.7MB of data every second, not having the right data solutions in place could be the most expensive mistake a business can make. HIPs can rapidly automate data and create a unified system that allows for the complete transformation of communication, ensuring the ability to get the right data to the right person at the right time to take advantage of critical insights and opportunities.

The property and casualty business is constantly evolving, with occasional seismic shifts that can transform the market – and require quick adaptation. The proposed insurance reforms in California are a prime example of such a transformative moment.

The proposed reforms could lead to much brighter days ahead for both insurers and California consumers. To navigate these changes and make the most of the opportunities presented, P&C insurers must review and adjust their strategies in the market.

This article explores three actions insurers can take to excel in this shifting landscape.

See also: Growing Number of Uninsurable Risks

Write the Lowest-Risk Properties in "At-Risk" Areas

Many carriers have likely put forward new or revised rate filings compliant with the April 2023 deadline of Regulation 2644.9 – and are awaiting California Department of Insurance (CDI) approval of their filing. Until these approvals are received, some carriers may hesitate to write in “at-risk” areas.

That may be an overly cautious approach. 

The reforms proposed would require admitted carriers to write policies in the wildfire-prone parts of the state – and do so for at least 85% of their statewide market share. For example, if a company provides 10% of policies across California, they would be required to provide 8.5% of the coverage in "at-risk" areas.

We expect smart insurers to race to write policies for the lowest-risk properties in these high-risk areas.

Why is this critical?  According to HazardHub data, approximately 25% of properties in the CDI-defined "at-risk" areas are likely profitable with current rates.

Insurers don't have to wait for their last filings to be approved before acting. Identifying these lower-risk properties in "at-risk" areas can give insurers a competitive edge while demonstrating goodwill through early compliance.

Acquire Advanced Analytics

The second action is understanding the importance of high-resolution analytics compared with the existing low-resolution analytics used by most insurers.

Traditional analytic approaches may no longer suffice as they are often (a) insufficiently granular, (b) rely solely on backward-looking data and (c) consider too few variables to assess and differentiate wildfire risk. They also tend to lump all properties into the same risk level across large census blocks or ZIP code regions.

Insurers that quickly embrace more sophisticated approaches will be able to differentiate risk using considerably more variables – and do so at the specific property parcel level instead of the census or ZIP code level.

Insurers that leap forward with advanced analytics will be able to accurately identify, price and manage risk at a granular level, giving them a significant competitive advantage.

See also: Data-Driven Transformation

Reset Strategy and Refile Rate Plans

The third action revolves around resetting strategy and reevaluating rate plans. This will be necessary to align with the proposed reforms and the changing market dynamics. Some steps insurers may take here include:

·         Assessing Current Market Share in the At-Risk Area: Evaluate existing market presence in California's "at-risk" areas, understanding the proportion of policies held in high-risk regions.

·         Defining Risk Appetite: Reexamine and define their risk appetite, recognizing that insurers will be able to charge premiums commensurate with risk.

·         Realigning Marketing Strategies: Adjust marketing strategies to effectively reach and serve homeowners in the state-defined at-risk wildfire regions.

·         Differentiating Offerings for Competitive Edge: In a market where not every insurer can compete solely on price, explore ways to provide added value, such as speed, ease of service and innovative coverage options. This differentiation will attract customers and help insurers meet their "at-risk" market share target.

As California moves forward with these proposed insurance reforms, it is clear that the market is entering a transformative chapter. Insurers must strategically adapt to the changes to excel in this evolving landscape. It is imperative for them to evaluate their market presence, redefine their risk appetite, revamp marketing strategies and explore avenues for differentiation. Embracing state-of-the-art risk analytics is essential to their success, equipping them with the capabilities needed to precisely identify, evaluate and price risk

The Equal Ventures Insurance Index is a quarterly summary of insurance equity performance and trends. We’re back to reflect on the insurance/insurtech ecosystem in Q3, which turned out to be a fairly meaningful quarter in terms of both price action and industry news.

Q3 Headline Summary

• In Q2, we noted that insurtech carriers rebounded and outperformed; this trend reversed sharply over Q3, as the insurtech carriers in our index were hit by rising loss ratios and the resulting deceleration in growth.
• Digital brokers once again underperformed their legacy peers, most likely dragged lower as the broader market traded sideways, signaling risk-off for most of small-cap tech.
• YTD, insurtechs and digital challengers have outpaced their legacy comparables, but the spread in YTD performance narrowed significantly in Q3 (so much for a quick path to valuation recovery…).
• Legacy brokers and P&C insurers were, on average, flat over the quarter, mirroring the broader market (and also masking larger continuing trends that we discuss in greater detail below).

Takeaways from Q3 Performance

Brokers continue to face a favorable environment as rates trend higher. The legacy names were flat on average in an otherwise choppy market. The index was dragged down by WTW, which reported lower margin and reduced guidance in Q2 earnings in part due to increased investments in talent — pointing to a broader trend in talent management that affects the insurance industry. On the digital challenger side, GSHD enjoyed another quarter of massive outperformance compared with the digital distribution peers we track it against and expanded its already outsized EPS and EBITDA multiples. The lead-gen and digital acquisition companies, on the other front, continued to underperform and for the most part traded sharply lower.

In the P&C carrier segment, observations from Q3 performance highlight a number of powerful and secular trends affecting the industry– chief among those trends is the extent to which property loss and cat exposure is pressuring margins and spooking investors.

Performance of the volatile insurtech carriers demonstrates this trend. LMND started the quarter strong, after catching a bid from its “synthetic agents” initiative as an opportunity to improve cash efficiency and accelerate growth. But the stock fell off a cliff (and hit all-time lows) after posting higher loss ratios driven by cat exposure in its homeowners segment and announcing reduced reinsurance coverage. As a result, the company announced it would slow growth while waiting out new rate filings/increases. This is likely the responsible thing to do but not at all what investors were looking for. The chart below shows the magnitude of the homeowners losses before/after cat losses.

See also: Lemonade's 'Synthetic Agent' Nonsense

HIPO performance tells a similar, if not exaggerated, story about destabilizing cat risks. HIPO reported elevated cat losses attributed to TX hail, and like LMND, made a commitment to slowing growth while rates increase. But shortly after earnings, HIPO abruptly announced an immediate and unprecedented pause of all new business while they “evaluate catastrophic risks, geographic diversification and enhanced underwriting…” Unsurprisingly, their stock fell more than 50% over Q3.

Importantly, the trend toward higher property cat losses was not limited to the volatile insurtechs. National carriers continued to exit the CA and FL markets, a trend that started making headlines in Q2. Alarmingly, a greater share of homeowners are forgoing insurance due to cost or accessibility, and there is an increasing reliance on public insurers of last resort (a trend aptly described by a CA insurance industry group last month as a “slow motion train wreck”).

But even outside of these hardest and most disrupted markets, carriers are grappling with increased nat cat event frequency and higher exposure. In the wake of the unstable and disorderly reinsurance renewals market earlier this year, carriers landed with less reinsurance coverage overall (and therefore more exposure to cat risks). At the same time, more conservative reinsurance contracts left carriers with higher retentions and more exclusions to unnamed storms and secondary perils. 2023 turned out to be a particularly active year for severe convective storms — exactly the type of unnamed, moderate-frequency events that carriers are more exposed to due to reinsurance constraints (and exactly the type of non-coastal property losses that HIPO and LMND referenced). Losses from named storms, on the other hand, for which reinsurers bear more risk, have (so far) been relatively subdued. As a result primary carriers faced losses in Q2 as much as 40% above normal, whereas reinsurers effectively enjoyed a windfall from higher rates and below-average loss. This explains both the volatility across P&C carriers in the index, as well as the outperformance of reinsurers.

See also: Insurtech Startups Are Doing It Again!

Beyond property insurance, in a market with reduced access to reinsurance and continuing claims inflation, specialty insurers and brokers continue to take share from the admitted market. Commercial carriers with outsized E&S lines continue to outperform carriers in admitted segments. KNSL, a pure-play E&S carrier (and therefore not part of our index) trades at >2.5x the earnings multiple of our legacy carrier index. As we’ve written about before, it’s up ~60% YTD through Q3 and points to a long-term shift toward specialty risk and niche commercial underwriting expertise. Commercial rates continued to increase in Q3, though the pace of rate growth decelerated sequentially, possibly pointing to more stability for carriers in the quarters ahead.

As a reminder, the purpose of this brief (and oversimplified) analysis is to share high-level trends for insurtech investors, and not to comment definitively on valuations or market expectations. That said, we believe it will be instructive to see how insurance equities fare in Q4, and we note that insurtech carriers appear to be struggling from similar macro trends that are affecting the P&C industry at large. Whatever happens, it is clear that controlling property losses will continue to take on increasing importance for both legacy carriers and digital challengers.

Across the last five years, about 40 million used car transactions happened each year (source: Bureau of Transportation Statistics). For most make, model and year combinations, a reasonable and current estimate of the insurance value of a vehicle in operation could be produced from a vehicle identification number (VIN).

This is not your claim adjusters’ Total Loss Value worksheet.  That would have mileage, condition, vehicle history, prior accidents and repairs, and like-for-like comparable vehicles in a local geography.

If you watch "Antiques Road Show" or even "Pawn Stars," you get the idea that everything has a value, that these can fluctuate and that the insurance value to replace something at market is a fungible concept, especially for things mass-produced – like a vehicle.

Before the government-imposed truth in labeling window sticker for vehicles emerged in the late 1950s, there was no ingredient label for what features were on a vehicle nor what a suggested retail price might be. As automotive retail strategies evolve, some manufacturers create many customer options while others produce more of a what-you-see-is-what-you-get menu.  

Before the days of the internet began making a direct-to-consumer order sheet a reality, companies would create their own estimates of what people want, then mass up incentives to clear the lots of anything not selling well. Those practices vastly changed due to the mall front business model of Tesla and the necessities of COVID shopping coupled with supply chain handcuffing.  A neo-epiphany was that you need way less inventory if you can build to order, and the order books were brimming with backlog.

See also: Auto Insurance in an Existential Crisis

The funny thing about the “have it your way” customer experience is that people don’t mind waiting a bit to get what they specifically prefer. The "car ordering" experience is completely different than the "car buying at a local lot" experience. The pressure on salespeople to clear the lot transfers directly to the shopper to “buy today what’s here” instead of buy when you are ready and “buy whatever you want,” where the value engineering of optional equipment and features can ruminate as shoppers sleep and as the showroom test what they can buy online.

This new behavior, coupled with supply chain and inventory constraints, may be a driving force behind recent year up-trimming and option-loading of vehicles entering operation as new inventory sales.  

While the COVID-crazy used car market still swells in lingering value, new sales have witnessed trim up-drift from the traditional production mix of basic, centerline and “premium trims” (source: Cloud Theory, Horizon Platform) to see popular vehicles with high choice configurations delivering many more premium “as built” vehicles today than in the past.

What this means is that if an insurer uses the manufacturer retail suggested price (MSRP) for setting an insurance value, it runs a high risk that the base MSRP will be several to tens of thousands of dollars below the “as built” fully configured total MSRP. And that means the insurer is undervaluing that vehicle from the start.

It gets worse. An object that starts with a higher value generally stays higher-valued than the rest of the production mix. So, if you insure at the base MSRP and then use a declining value forecast, the higher total MSRP vehicles will float above that for a decade. It’s tough to sell insurance for a $65,000 vehicle while charging like it is a $48,000 vehicle.  

While some may argue that at those cost levels any vehicle may only be short $100 a policy period, do remember that when mass produced that figure is multiplied by tens or hundreds of thousands of vehicles in any single model year. As model years accumulate, this undervaluation accumulates, as well. Then, as you factor in the general up-swelling of use vehicle values, a combo effect can kick in, where the accumulation promulgates into a new plateau of actual costs that are uncollected as exposure for the actual cash value of these subjects-at-risk.

See also: Setting Record Straight on Auto Claims Severity

The hangover effect of using an embedded but outdated and under-segmented vehicle valuation predictive model approach from years past is catching up with auto insurers.

Milestones have been passed: window stickers (1958), 17-digit VIN standardization (1981), auto insurance sold on the internet (1997) and ubiquitous sales information for vehicles transparently available for all the inventory on the internet (last 10 years). These created a new way of working with digital data. Perhaps leaving behind static predictions that underperform and using dynamic and continuous data with market savvy valuations is what auto insurance need to do personalized risk-based pricing right.

While there are a lot of reasons that help to explain why vehicle values went up, there are not a lot of good ideas of what will make them go down a lot anytime soon. Adopt and adapt new thinking for the value of the subject at risk for risk-based pricing in auto insurance.

Best practice in data and analytics -- Don’t predict what you can describe.

Rob Galbraith is the founder and CEO of Forestview Insights, an independent consulting, research and training company focused on helping organizations build a culture of innovation. He is the author of the international bestselling book "The End Of Insurance As We Know It" and a popular keynote speaker who has shared his unique insights at numerous events around the globe. Galbraith has over 25 years of experience in the financial services industry in a variety of leadership positions. He is a recognized thought leader on P&C insurance who is a frequent media contributor and well-known industry influencer. He holds a master's of science in insurance management from Boston University and a bachelor of arts in economics from Michigan State University. He has earned the CPCU, CLU, and ChFC professional designations.

Insurance Thought Leadership:

It has been almost five years since your book "The End of Insurance as We Know It" was published. What are some of your forecasts that have come true, and are there any that did not come to pass as expected?

Rob Galbraith:

A few months ago, I connected with a startup founder who recently discovered my book, and he exclaimed, “Everything you predicted came true!” I so wish that was the case! My honest assessment would be that the book's core argument remains true – that emerging technologies are coalescing with large amounts of funding and newer, tech-savvy generations to change the insurance industry. I also argued that artificial intelligence would quickly become the most transformational emerging technology, and we see that a lot more clearly today with conversations around generative AI and the ethical applications of AI to remove bias in the industry.

On the flip side, while numerous headlines and a lot of buzz have been generated, the speed and scale of the predicted transformation have been slower to materialize than I anticipated. This has been less due to the challenges of promoting innovation and change with traditional insurance firms – always a big hill to climb - and more due to the relative struggles of outside startups to grab market share against the incumbents and challenge their supremacy. I think there has been more success in specialty and niche segments that were not covered or underserved but less disruption in the core P&C, life and health markets than I originally anticipated. I also foresaw a slower adoption of blockchain, but even with more tempered expectations, I am surprised how little to date blockchain specifically and Web3 more generally have affected the industry. I applaud the steady progress of the RiskStream Collaborative in this area, where many other initiatives have faded.

ITL:

What is an area (or areas) that you believe remains untapped/unfulfilled/overlooked for the promise of innovation in insurance?

Galbraith:

I would categorize investments in innovation based on an insurer’s income statement. Most innovation efforts have focused on increasing revenues through new product offerings and new (digital) sales channels. Another area of significant investment has been reducing expenses and streamlining key processes through systems modernization efforts, automation and low-code/no-code platforms, among others.

What is unique at this moment in time is that companies can simultaneously reduce operational expenses and improve customer experience. Usually, there is a trade-off involved where expenses increase as firms work to improve customer service.

I see a lot of underinvestment in the claims space. Outside of implementing new fraud detection solutions, most firms have shied away from large investments to help better manage loss payouts and reduce loss adjustment expenses. I also think more investment is needed in risk prevention efforts – these tend to have a longer tail to see meaningful benefits in the form of reduced losses, and the ROI is highly uncertain with longer payback periods, so these projects lose out to other efforts with a more concrete return. I also think billing and ancillary functions such as regulatory compliance, premium audit and subrogation haven’t received enough attention.

ITL:

What do you see as the biggest obstacles to insurance innovation, and how would you recommend overcoming them?

Galbraith:

My first takeaway is that innovation is a discipline, not a side show—it should be seen as equal to more established disciplines such as claims, underwriting, actuarial science, IT, HR, etc. The key difference is that it’s a much younger discipline that most firms have focused on for less than two decades, so best practices have yet to be evaluated, well-documented and widely disseminated. Professional designations and continuing education requirements exist for agents, brokers, claims personnel, underwriters, actuaries, IT, HR, etc. These serve as ways to establish a community of practice, and when professionals move from one company to the next, the best practices of that discipline remain in place to guide people. This is not true yet for innovation—each organization approaches the topic quite differently. I help companies assess their innovation efforts and conduct training and workshops to make them more effective based on my past successes and failures.

My second takeaway is that three forces drive innovation: top-down, bottom-up and outside the organization. Top-down efforts, by definition, have senior-level support, but if they aren’t effectively tied to a strategic road map that makes sense, they won’t last beyond the leader who initiates them. Bottom-up efforts are plentiful but must fight hard for attention and resources, are often scattered and incompatible, and usually result in incremental innovation with marginal benefits. Outside forces (technology, the economy, shifts in the market, regulators, etc.) can force companies to respond through innovation, but the resistance to change is high. Reactive innovation efforts struggle to find internal champions and gain enough traction to propel them over the finish line, and the benefits are muted because other competitors have also adapted. Ideally, organizations rely on a blend of “changemakers” incorporating all three forces to have a balanced approach and allow for building a culture of innovation where change is expected and embraced.

ITL:

This year’s launch of Chat GPT has captured the imagination for the potential applications of AI to various businesses, as well as thousands of actual applications. How would you grade insurance efforts to leverage AI?

Galbraith:

On a scale of 1 to 10, I rate insurance efforts at a 4, which may be generous. The challenge isn’t the lack of adoption but rather the speed of adoption. AI is a critical cornerstone technology full of possibilities and challenges, and this year has proven that it continues to evolve rapidly. Traditional insurance organizations generally wait to deploy technologies once they have become stable and proven, but delaying efforts to incorporate AI in a wide range of use cases will create a competitive disadvantage.

Today’s biggest challenge is the hard insurance market, and while there is a confluence of forces that have led to current conditions -- including inflationary pressures, rising interest rates and climate change – clear solutions aren’t as obvious. Raising rates, restricting risk appetites and reducing expenses are insufficient on their own. To achieve profitability, risk selection is paramount – and the only way to improve on current methods is to acquire new, highly granular data. Whether this data comes from IoT-enabled sensors or unstructured sources such as handwritten notes, images or video, artificial intelligence is essential to find meaningful patterns in these rich and previously underused data sets and guide our actions. We have seen the power that AI has in Big Tech and need to accelerate our efforts as an industry to harness its full potential.

ITL:

You recently presented a webinar with The Institutes on insurtech. What are some of the lessons you hope attendees will take away from the course and bring back to their organizations?

Galbraith:

There are three core lessons that I hope attendees leave with. First, we live in an amazing time characterized by rapid technological change. I use an acronym to describe what I call the SCALED emerging technologies of today: Sensors, Cloud, AI, Localized knowledge, Efficiencies and experiences, and Digital distribution and communication. We typically learn about each of these technologies in isolation, but it’s important to see the bigger picture and understand how they support each other, how they interact and the amazing new capabilities and possibilities they unlock.

The second lesson is the importance of seeing innovation as a discipline, as I previously mentioned. Organizations establish and adopt best practices such that when professionals move from one firm to another, the essence of the function is the same, and each has a different “flavor” that makes up its “secret sauce.” By contrast, innovation is an immature discipline that has only been rigorously practiced within insurance organizations for a decade or two. Each firm varies widely in how they approach innovation, yet this discipline is critical for translating the SCALED emerging technologies and applying them in the best contexts for insurance organizations to gain a competitive advantage and better serve customer needs that are being shaped by leaders outside of insurance such as Apple and Amazon.

The third lesson is that innovation isn’t just the domain of a small group of specialists but your entire organization. As the saying goes, “None of us is as smart as all of us.” Innovation is about novel ways of doing business, not merely incorporating a new piece of technology to do business the same way as before. Title and rank matter for access to resources, but there is no monopoly on good ideas. Other necessary ingredients are cognitive diversity to avoid groupthink and psychological safety to speak your mind without fear of repercussion. Failure should not just be accepted but encouraged. Innovation is ultimately about experimentation – learning what works and what doesn’t when there are no clear answers because no one can see the future perfectly. To succeed in the long run, you must build a culture of innovation that goes beyond individual leaders or projects and is ingrained within your organization.

ITL:

The Institutes is leading conversations on Predict & Prevent, the idea of using technology to predict potential losses and prevent them before they can become a claim. What are your thoughts on this shift in how insurance can serve its customers?

Galbraith:

One of the most common questions I receive from industry professionals is, “when is risk prevention going to become a reality?” There was a lot of hype in the mid-2010s about self-driving cars and IoT devices and shifting from a risk recovery to a risk prevention business model for insurance organizations, yet the reality is we have yet to move the needle much. I am optimistic that we can achieve this vision in the long run, and I love that The Institutes is taking a leading role in helping shape this paradigm. It’s a major undertaking, not just from a technological standpoint, but reinventing entire processes and business models that are hard-wired into our industry. In situations like this, I remind myself of the 10/10 Rule: it takes 10 years for emerging technologies to go from conceptual to useful and another 10 years to reach mass adoption.

Think of the social, psychological and economic benefits of loss prevention: it is extremely powerful and a vastly superior alternative to the reactive approach that has characterized insurance since its inception centuries ago. In collaboration with partners, loss prevention is a focus of our industry today in many contexts, such as seat belts, worker safety, building codes, etc. With emerging technologies and innovative approaches, the insurance industry can go far beyond our current capabilities and, ideally, be seen more as true collaborative partners that help people and organizations rather than a source of friction.

ITL:

Thank you, Rob. Are there any other issues or topics that you think we should be paying attention to?

Galbraith:

I’m curious where people assess the state of insurtech in the context of the Gartner “hype cycle” today. With the current hard market challenging profitability across the board for insurance organizations and the challenges that many startups have faced, are innovation efforts moving to the back burner? Are we currently in the “trough of disillusionment” today, and what does the slow adoption and acceptance phase look like? Will a startup competitor like Lemonade or Hippo truly challenge the incumbents and shake up the top 10 in our industry? Some have argued we have not seen this confluence of events – inflation, global instability, high interest rates, large losses – in 40 years or more in the insurance industry. Can innovation lead us back to profitability? Or will market conditions and challenges with availability and affordability persist?

Since diving into the insurance sector right out of school 36 years ago, I've enjoyed watching the so-called experts predict the demise of the local personal lines property/casualty insurance agent across the years.

There have been several waves of direct-to-consumer (D2C) personal lines P&C insurance provider start-ups throughout that timeframe, and each time we've been left dodging all the subsequent roadkill (or soon-to-be-roadkill among the most recent wave of insurtechs). The next wave will be generated by the "experts" who jump onto the artificial bandwagon.

There are only a couple honest D2C successes over time, and it took a long time for each...and a lot of patience and capital to support the steady growth.

Take Wawanesa. It started in 1975 in San Diego with the support of the Canadian parent, stayed focused regionally and expanded into Oregon around 30 years later. To this day, it only sells policies in California and Oregon. It's an absolute gem of an insurance provider. There's also The General (originally Permanent General) that was launched by Brian Brown in Nashville. And obviously, we have GEICO and Progressive.

GEICO nearly failed in the '70s but was rescued by the Oracle of Omaha, starting in 1976. Per its website, Progressive shifted "focus from being an agent-centric company to building a leading consumer brand" in 1988. With that strategy shift, a legendary leader and a bunch of brilliant, motivated people like Alan Bauer, Progressive thumbed its nose at slow, steady growth and began its meteoric market share grab. (For more on that story, please take 5 minutes to read my all-time favorite article on auto insurance, "Sex. Reefer? And Auto Insurance.")  

But that's about it! Pretty much all others that set out to "crush" the antiquated and bloated local agency model in the last four decades have failed. Fireman's Fund, Great American and Direct Response tried in the '90s, eCoverage in the early 2000s. It even took out a full-page color ad in the Wall Street Journal proclaiming "The Industry is Toast." A year or two later, it shut down and sold its technology to GMAC Insurance for practically nothing.

Oh, and don't forget about Esurance! Did Erin Esurance ever get her loss ratio into the double digits? The insurtech movement burped out "agent killers" like Hippo, Root and Lemonade (to name a few), which all had seriously broken IPOs and, now, rapidly declining market caps.

Now, everyone should get ready (and be very skeptical) of the fat cat consulting firms roaming corporate insurance company hallways who will wave the artificial intelligence flag as the next thing that will no doubt exterminate the local agent. These are the folks who have LinkedIn profiles with "MBA" after their names. because they think it makes them look smarter and more important. like a doctor or professor. (Not in my book....)

The consultants will bash the local agent model and issue recommendations to ditch it and go direct. No consultant can make a living by saying, "You're doing a great job, keep up the great work and don't change a thing!"

Those office water walls and seafood tower dinners aren't going to pay for themselves!

Why does the local agent continue to dominate personal lines auto and home distribution? And why will they become even more important in the artificial intelligence age?

Simple...it's the consumer, stupid! Consumer behavior hasn't changed much when it comes to shopping and buying auto and home insurance. Sure, you have the non-standard consumer who will buy from anyone and the ultra-price-sensitive consumer who will move providers for a couple bucks of savings. But the vast majority of consumers want a local insurance professional who can expertly guide them through the complex buying process to make sure proper coverages are established and be nearby when "the promise" is needed in tough times.

As artificial intelligence continues its rapid rise, I expect consumers will increasingly become skeptical when it comes to insurance, and it will be difficult to trust the machines with protecting their most prized possessions. And forget about trying to have an actual meaningful conversation with a real human being to determine the best coverages for their individual situations.

To you insurance carriers: Don't discount the plus of having boots on the ground to verify reality when needed (this will become increasingly difficult from afar).

The result: Consumers will lean even more into the local agent -- someone they can actually see in person (if they want to) and know is real and nearby. You can't do that with artificial intelligence.

The "experts" will argue that robots residing on corporate servers can do a much better job for consumers than local agents! Please do a focus group before deciding to follow that flawed thinking.

With luck, some carriers will shift gears and begin to invest in agents' compensation, education, development and operations. Those that do will dominate the passing lane on the artificial intelligence superhighway.

To local agents: Sorry, but it's likely you'll all continue to be dissed just as you have been throughout the past. Don't listen, and don't take it.

All of the local insurance professionals I know have tough skin. Rock on like Twisted Sister! "There ain't no way we'll lose it, this is our life, this is our song, we'll fight the powers that be, we're not gonna take it."

The headline in Business Insider couldn't be starker: "Insurance Companies Have Discovered Devious New Ways to Rip You Off." And I think it should be a call to action.

People in the industry understand the need to sharply raise rates for auto and homeowners insurance. Replacement costs are way up, climbing far faster even than the lofty inflation rates of the past two years. The addition of sophisticated electronics in cars and the transition to electric vehicles increase repair costs, too. Frequency of claims is up, as well, and will likely keep increasing, especially for homeowners, as the warming climate causes more and bigger storms and wildfires. 

As costs rise, premiums have to, too. Insurance is a business, not a hobby.

But consumers aren't in the industry and aren't exactly sympathetic to insurers' problems. They just see rates soaring, and they're angry. They will now generate as much pressure as they can on state regulators -- whom they elect, either directly or indirectly -- and we can expect to see a lot more headlines about "devious" insurers.

What to do?

While nothing will eliminate the problem, I think we can take at least some of the sting out of it by doing three things.

First, we need to communicate, communicate and communicate with customers so they understand why rates have to rise. Then we need to communicate some more.

Most of the burden will fall on agents and brokers, who need to explain, in detail, why costs have soared and, thus, why premiums need to climb, too.

Agents and brokers should also be crystal clear about changes in coverage. Many customers are inclined to scale back to reduce premiums, yet will blame agents and brokers or carriers if they have a loss and find that an exclusion they've forgotten about means it isn't covered. Agents, brokers and carriers all need to get customers to acknowledge what they've decided not to cover and to remind them from time to time. A surprised customer is a furious customer. 

It'd be great if the carriers would lean into educating the public about rising costs, but they seem to view talking about the need for rate increases as an opening for competitors to brag about lower prices. So we'll likely just keep being inundated with geckos, emus and all possible permutations of Flo and Mayhem as the carriers hammer away at improving brand recognition. Our colleagues at the Insurance Information Institute are doing yeoman's work on behalf of the industry, but they can only reach so many people through their comments in the media.

Carriers should, at least, help agents and brokers work with customers on how to reduce premiums -- being as specific as possible for individual customers about how they can diminish, say, wildfire risk or the potential for water damage. If the insurance industry demonstrates good will, customers will be at least a bit more sympathetic about price increases. 

Second, we need to be far more aggressive about lowering costs, as a way of diminishing the need for rate increases. Geico just announced that it is laying off 2,000 employees, or about 5% of its workforce, and Farmers Insurance said in late August that it was laying off 2,400 people, or some 11% of its employees. But there are loads of other ways that insurers can become more efficient, too.

Matthew Grant, CEO of Instech, said in a recent conversation that insurtechs have developed lots of ways that carriers can cut costs, and carriers see clear benefits, but "there wasn't really a burning problem, so are you going to make the effort to bring something into the company?" He said insurtechs' cost-cutting innovations always seemed to be "No. 11 on the list of boards' top 10 priorities."

Cost-cutting should move way up those lists -- and carriers should make sure customers know about it, so, again, they see the insurance industry working on their behalf to minimize price increases as much as humanly possible.

Third, the industry should move as swiftly as possible toward the "Predict & Prevent" model and away from the traditional "repair and replace." We have all this data about where the risks are. On the theory that the best claim is the one that never happens, let's use that data to help customers prevent losses, rather than just help them recover after something goes very wrong.

We've published extensively on the topic at ITL, and I'd encourage you to check out some of the very smart pieces. Here is a smorgasbord, based on a search for "Predict & Prevent" on the ITL site. If you have time for just one, I'd suggest this webinar I hosted with Pete Miller, the CEO of The Institutes; David Harkey, president of the Insurance Institute for Highway Safety; and Roy Wright, president and CEO of the Insurance Institute for Building & Home Safety.

Nobody wants to buy insurance, but everybody wants protection.

With all that said, I think the piece in Business Insider is, well, hardly an example of journalistic excellence. While it's posted on the site as a news article, a note at the end describes it as part of an initiative that provides "thought-provoking perspectives, informed by analysis, reporting and expertise." In other words, opinion. 

The piece was written by a professor at an Australian university who wrote a book a few years ago about how businesses are, to quote the subtitle, "extracting data, controlling our lives and taking over the world." The closest the piece comes to quoting someone as saying something negative is a paraphrase of Dame Inga Beale from a talk she gave, and there's no link to the talk, so I can't see precisely what she said or the context for her words. 

The article makes claims that will strike any student of business as silly. Insurers are dinged for focusing on the lifetime value of customers -- something every sentient business does. The article also sneers at usage-based insurance as somehow unfairly intrusive and at how insurers "might look at your home's roof using drones and automated image analysis, or where you're driving based on data from a smart device in your car."

What's devious about any of that?

Where the article delves into issues that could be legitimate concerns -- using AI for "price optimization," based not on risk but on customers' ability and willingness to pay, and "claims optimization," based not on what should be paid but on how much customers complain if claims are denied -- the writer cites no evidence and quotes no one. He also doesn't try to quantify the extent of these alleged practices. He just asserts that insurance companies are guilty.

So, while Business Insider is generally a legitimate publication, I don't take the complaints at all seriously. 

But my analysis of the article doesn't matter. Neither does yours. What matters is that the article is out there and that a lot more like it may be in the works because consumers are angry about soaring premiums.

We as an industry need to change the narrative as aggressively as we can, or we'll let this public relations crisis define us.

Cheers,

Paul