By Russ Banham
Risk Management
Among the hundreds of startups in the burgeoning field of insurtech—technology innovations designed to improve the insurance industry—are close to a dozen peer-to-peer (P2P) companies looking to compete directly with traditional insurance markets. While only one such company, Lemonade, is currently licensed to do business in the United States—and only in New York state—other P2P insurance companies have sprung up around the world.
Analysts have mixed opinions about the potential for the P2P companies to make market inroads. Some predict the startups will take significant market share from insurers, while others remain dubious. Most believe the P2P model appears to have a lot going for it, though, insofar as it could improve risk management, decrease the incidence of insurance fraud, and lower the cost of insurance.
So what is a P2P insurer? In many ways, it is a throwback to when groups of people came together to insure each other’s risks. This sounds a lot like mutual insurance, but one key distinction is that the policyholders in a P2P group do not own the company absorbing their exposures. The model also borrows from traditional captive insurance mechanisms in that the insurer bears either very little or no insured risk, ceding the losses to reinsurers.
Lemonade is a case in point. The personal lines insurance company underwrites renters and homeowners insurance. Premiums collected from group members are pooled to pay first-dollar losses. Beyond this threshold, losses are absorbed by its global reinsurance partners, which include Lloyd’s of London, Everest Re and Berkshire Hathaway’s National Indemnity.
Other than its unusual name, Lemonade is especially unique in how groups of policyholders are formed. When signing up to buy insurance, the applicant selects a particular charity. Others who choose the same charity are assembled as a group of policyholders for insurance purposes. When losses during the policy period are low enough to result in unspent premiums, this capital is donated to the charity.
Other P2P companies like Germany’s Friendsurance, which is a broker and not an insurer, have a slightly different model. Groups of policyholders are still formed, but they comprise people and their selected friends who join together to insure each other’s personal liability, legal expenses, automobile and household risks. After claims are paid out, any unspent premiums are returned to group participants.
Both cases depend on a form of peer pressure. Lemonade’s policyholders theoretically will try to be vigilant about their risks to preserve capital for a charity they believe in. Friendsurance’s policyholders will be equally alert to keeping losses down for the group, as their friends’ premiums may otherwise increase the following year. Participants in both companies are also theoretically less likely to commit insurance fraud for the same reasons.
For now, these companies have concentrated their focus primarily on personal lines and some small commercial uses. All are expected to broaden their offerings if the P2P concept catches on, possibly selling more specialized lines of commercial property and casualty insurance like workers compensation and employment practices liability.
“I believe the model has legs,” said Steve Kauderer, a partner in the financial services and insurance practice of management consultancy Bain & Company. “Right now they’re in the spectrum between personal lines and small commercial, but we expect them to move up in complexity.”
If he is right, corporate risk managers eventually may have another risk transfer option—one that may be more risk-sensitive and cost-effective than traditional insurance, thanks to the aforementioned peer pressure and the digital cloud-based solutions these startups use to underwrite, market and process claims more efficiently.
“If you are able to create a tight-knit group [of companies] in a community that is set up like a captive but is managed by a P2P entity that handles all the underwriting and claims administration baked into a competitive premium, I can easily see the model working for smaller businesses,” said Anand Rao, principal in the insurance advisory services practice in PwC’s Analytics Group. “I’m a bit more skeptical as you get to larger companies with more nuanced risks.”
Slow, Steady Growth
Analysts are not exactly calling P2P models the insurance equivalent of Uber, aimed at displacing the traditional market, although some P2P companies might like buyers to believe this.
In numerous articles and blogs, Lemonade’s Chief Behavioral Scientist Dan Ariely routinely criticizes the insurance industry for not returning unspent premium dollars to policyholders. He and other Lemonade leaders (who were unavailable for comment) insist this money belongs to policyholders and not to insurers as underwriting profit. Not that Lemonade gives the unspent premiums back to policyholders either—it deducts 20% of policyholder premiums as its fee and the rest flows instead to charities.
Ariely is the James B. Duke Professor of Psychology and Behavioral Economics at Duke University and the founder of the Center for Advanced Hindsight. His appointment by Lemonade emphasizes the behavioral underpinnings of its P2P model. The traditional insurance industry, in his view, has little to no appreciation for human behavior.
“We’ve spent recent years deepening our understanding of honesty and trust, and our conclusion is that insurance is crying out for a makeover,” Ariely said in a statement. “Lemonade aims to reverse the adversarial dynamics that plague the industry, transforming both the economics and experience of insurance.”
Lemonade began with renters and homeowners insurance products in New York when its license to sell insurance was approved by state regulators in September. The company has filed for licenses in 46 states and the District of Columbia and hopes to become available to 97% of the U.S. population in 2017. It also reportedly plans to underwrite commercial lines insurance at some point in the future. “Staffing Lemonade with bots instead of brokers allows for rapid expansion,” Shai Wininger, president and cofounder, said in a statement.
As the first P2P insurance company in the United States, Lemonade’s progress is being closely watched by the array of similar entities that have formed around the world. This long (and growing) list includes Guevara, a P2P brokerage model in the United Kingdom with a structure and value proposition analogous to Friendsurance’s; P2P Protect Co., Ltd. in Hong Kong; PeerCover in New Zealand; Riovic in South Africa; insPeer in France; První Klubová pojišťovna in the Czech Republic; TongJuBao in China; Huddle Money in Australia; and Besure in Canada.
It is difficult to provide an accurate prognosis for these startups. Fred Eslami, a senior financial analyst at A.M. Best, said that the rating agency has had difficulty communicating with the companies. “They’re pretty reticent when it comes to offering information,” he explained. “They don’t share too much with the outside world, so we don’t really know what their market share or premium levels are.”
Lemonade did reveal that it had sold $14,302 in gross written premiums in its first 48 hours and a total of $179,855 in gross written premiums in 2016. While not earth-shattering numbers, the company has nevertheless attracted substantial venture capital interest, announcing in December 2016 the closure of a $34 million Series B funding round. The additional capital augments the company’s total funding for full-year 2016 to $60 million. The money will go toward “rapid expansion in the coming months,” Lemonade said.
How rapidly will Lemonade and other P2P companies grow? In terms of writing large commercial lines, Eslami said he would be surprised if that occurred in the next five years. Kauderer predicted the pace will be “slow and steady.”
Kaenan Hertz, head of insurance innovation and fintech at EY, likened the current status of P2P companies to the internet in the late-1980, where what began as niche applications evolved into the sprawling network of today. He believes P2P could have a similar trajectory. “Conceptually, if you are bringing together people with a common cause—people who know each other or have similar opinions and feelings—this will compel them to maintain discipline when it comes to their respective risks, ultimately reducing the exposure to loss,” he said.
Are Companies People Too?
This argument makes sense when it comes to people who are friends or share an interest in a particular philanthropic organization, but what about companies and their risk managers—can peer pressure do the same for them?
“P2P may work well for commoditized personal lines business, but I’m a skeptic when it comes to its value in commercial lines—even small commercial,” said Joe Calandro, managing director of PwC’s insurance practice. “It’s one thing for a policyholder who has picked your charity to defraud the group; it’s quite another for an employee in a large company to embezzle money. I don’t see how the P2P relationship-oriented model will have any bearing on this risk.”
Indeed, if the threat of arrest and confinement is not dire enough to keep someone from defrauding his or her employer, what effect, if any, will peer pressure have? And without peer pressure, the prospect of risk containment fades. “If the group experiences greater claims frequency and severity, companies will drop out of the community,” Rao said. “Since insurance is all about spreading risk and not concentrating it in a small group, there may actually be a greater risk of loss.”
With regard to Lemonade specifically, Rao questioned whether people sharing an interest in the same charity have the same behavioral characteristics as, for example, people living and working in a small community. “People really know each other in a small town—that’s peer pressure,” he explained. “Social and physical values define the community. I don’t see this in Lemonade. And I have trouble seeing it in a digital medium, where there are no real face-to-face human interactions.” In general, he believes any community connected solely online is going to be desensitized to mutuality.
Hertz raised another concern—the possibility that policyholders can collectively perpetrate a large-scale insurance fraud. “You could conceivably end up with a pretty tight group of people in a P2P group who are corrupt and able to game the system,” he said. “A P2P group of risk managers from different companies could decide to adjudicate the group’s claims in such a way that more losses are passed on to the reinsurers, forcing them to carry the load.”
He added, “The behavioral economics could backfire.”
Taxing Matters
One factor that will test the mettle of P2P companies like Lemonade is the tax code in New York. Right now, group policyholders whose premiums are donated to a charity do not receive the usual tax deduction they would for a more standard charitable contribution. Rather, Lemonade gets the tax deduction. That is because the state may perceive the deduction by a policyholder as a rebate in disguise.
Assuming Lemonade underwrites commercial lines insurance using the same charity concept, companies may similarly lose out on the tax deductibility of the donation. This also presumes that businesses will want to make a donation to a charity in the first place, given the reputational consequences of choosing a philanthropic organization that may upset customers across the political spectrum.
The premiums paid to Lemonade are fully deductible as a business expense, which is not the case for captives that do not underwrite third-party insurance. This is just one factor that makes P2P groups a competitive alternative to traditional single-parent captives.
“Instead of managing a captive from an insurance administrative standpoint, the P2P entity effectively handles this, liberating risk managers to focus on risk mitigation, putting in place better processes and technologies to minimize loss, and becoming better business partners to the rest of the organization,” Hertz said.
Tax deductibility is less of an issue with captives that underwrite third-party business, such as many industry-owned captives and risk-pooling arrangements like broker Marsh’s Green Island Reinsurance Treaty, in which single-parent captives share their loss experience by transferring a portion of their risk in exchange for a percentage share of the risks of other pool members.
Down the Line
Assuming the P2P model takes root, there is no reason for traditional insurers not to form their own P2P groups for buyers of personal and commercial lines products. They also may invest in or even buy P2P startups. Many insurers have an enormous volume of capital sitting on the sidelines and several have formed venture capital funds to invest in novel insurance concepts.
“Undoubtedly many traditional carriers are looking at these guys and figuring out whether to buy them wholly, invest in them, start their own P2P groups, or just wait and see how the dust settles,” Kauderer said.
The bottom line is that P2P insurance is just another way to transfer risk. While it might excite younger, first-time buyers, it is ultimately a new combination of old concepts—captives and mutual insurance. This may also offer buyers more choices in how to transfer their risks, and greater market power in turn. “Some P2P entities will survive and flourish, putting pressure on traditional carriers to compete,” Kauderer said. “Competition is always a good thing.”