Etherisc: A Decentralized Insurance Protocol

Etherisc is developing an insurance protocol based on Ethereum smart contracts. The ecosystem involves an array of service providers launching and selling various types of insurance, backed by on-chain and off-chain risk capital and reinsurance capital. The DIP token is used by service providers, who must stake an amount worth 15 - 25% of their monthly revenue.

Etherisc is developing a decentralized insurance protocol intended to support the emergence of a wide range of specialized insurance products across a variety of markets. The distributed, smart contract-based system is conceived of as a platform dApps can build on and extend with smart contract libraries and risk models that enable products like flight insurance or localized weather insurance. The aim is for the products to be less expensive and less subjective than products from traditional insurance companies, while also being more transparent and responsive. Several insurance products are being launched on Etherisc, and a live MVP insurance product insuring against flight delays has issued ETH payouts to several policy holders. The DIP token functions as a staking requirement for anyone collecting fees on the platform.

What the Decentralized Insurance Protocol Intends to Accomplish

The modern insurance industry is built around three core elements:

  1. Calculated expected value of the risk: refers to the anticipated amount to be paid out over an identified time frame. If crops are being insured against failure, what do historical records and statistical tables indicate will be the expected number of failures in a given season or year? This informs premiums charged by the insurer.
  2. Capital cost (or reinsurance) for long tail risks: effectively are a reserve in the event that a 100 year drought sends crop failures out to multiples of anticipated amounts. Does the system have the capital reserves to cover this eventuality as well?
  3. Transaction costs: are the incidental costs associated with normal business operations, and are also related to the complexity and customization required per product.

If the above represent the core elements of an insurance operation, in practice, most of the world interacts with a system where two additional elements have become commonplace:

  1. Administration: refers to the general overhead of the large insurance companies that are required to operate their system, the costs of which must be built into the premiums individuals and businesses pay for coverage.
  2. Shareholder return: must be factored into premiums and decisions regarding payouts, at least to the extent that insurance companies have shareholders expecting a return on their own capital.

Etherisc proposes a decentralized protocol upon which anyone can launch smart-contract based insurance products. They contend that a blockchain-based platform has the potential to decrease transaction costs, administration costs, and remove the pressures of shareholder return, while at the same time enabling new parties to contribute both models for calculating the expected value of risk and capital for long-tail risks. Smart contracts can automate premium collection and payouts, while also making obligations more transparent. A protocol that standardizes how insurance products are presented, structured, and managed creates additional opportunities to reduce inefficiencies and administrative overhead on a corporate level, enabling the further reduction premiums. Beyond this, creating a protocol-based system, where token holders effectively replace shareholders, will remove the need to provide a return to capital, allowing a further reduction of financial obligations that filters back to premiums.

This approach lends itself in particular to a focus on parametric insurance markets, where outcomes are generally quantitative ones that are informed by sensors or other automated means that are able to communicate with smart contracts. An example can be seen in the hurricane insurance program, where payouts are based solely on recorded wind speeds, in flight insurance programs that measure delays, or in crop insurance projects based entirely upon rainfall amounts. This type of insurance generally relies on objective indicators and has payouts not determined by estimated loss, leaving insurance companies less room to fight payouts. Etherisc also argues that traditional insurance against loss will nonetheless benefit from a system that at the very least, reduces the administrative costs of managing capital.

In addition, compared to typical insurance companies that often work to avoid or reduce customer payments, the more transparent nature of a smart-contract based system should reduce both time spent disputing and disappointment with insurance policies, due to less opaque obligations. The overall result is imagined as a new type of insurance company that will be able to honor justified claims and rapidly settle outstanding issues, all while dedicating a considerably smaller percentage of premiums to administration and overhead. When Etherisc claims that a blockchain-based insurance system could be operated for an order-of-magnitude less than a traditional insurance company, the scale of the opportunity can be appreciated.

How the Insurance Protocol Works

The system involves an array of on-chain actors that have direct counterparts in the traditional insurance industry.

Keepers: keepers package and offer insurance products based on smart contracts. Etherisc imagines these could be entrepreneurs, data scientists with new risk models, or traditional insurance providers offering new products. Keepers determine the premium and payouts of their products. They are also responsible for determining whether policyholders are allowed to buy policies e.g. farmers must actually plant crops to qualify for crop insurance.

They are also considered responsible for recruiting an array of other service providers as needed, discussed below. Keepers can fully integrate these functions or rely on partners.

  • Underwriters: Underwriter evaluate risk models for pricing policies, determining the premiums and payouts to keep policies solvent.
  • Oracles: Verify off-chain conditions that should trigger payouts. In cases of parametric insurance, systems like sensor networks or information sources could serve as oracles, or trusted third-parties could serve as final verifiers of payouts.
  • License Providers: Keepers ultimately choose whether to operate with a license. Etherisc imagines that established insurance companies could earn fees for ‘renting’ or allowing others to operate under their license(s) in a particular jurisdiction. As with keepers, fees are established by the provider, encouraging competition between providers.
  • Claims Adjusters: In cases where loss needs to be estimated, claims adjusters would be responsible for providing the necessary adjustments to payouts.

Risk Capital: Keepers may collateralize the product with their own capital, or acquire additional backing for their product either through collateralizing a risk pool and selling tokenized fractions of it, or acquire reinsurance through a licensed reinsurer.

Relayers: Relayers are responsible for connecting users and the products that Keepers bring to the market. These website would advertise insurance, likely from multiple Keepers. Relayers are imagined as helping police against bad products by choosing not to list them, for example, by choosing to only list products with insurance licenses, though Relayers can list any product they want. Keepers can also establish themselves as their own Relayers, offering a website on which people can sign up for policies. Flightrisk, operating today, is example of a product choosing to advertise on its own site, though presumably eventually consumers will need insurance product aggregators to discover and compare insurance products. Relayers establish their own fees for listing insurance products.

Etherisc Cryptoeconomic System

Etherisc is structured around what it calls a ‘Two-Fold Approach,’ reflecting the structure of the ‘Decentralized Insurance Foundation’ and the multiple for-profit commercial entities unique to a jurisdiction that form companies such as Ethereisc Holding AG, Etherisc US, or Etherisc MT. These jurisdictionally specific entities are in turn controlled by the Decentralized Insurance Foundation. The jurisdiction-based entities serve as the local repository of contracts and provider of platform services, allowing and encouraging new projects to join to access the smart contracts and services within the protocol. This structure should allow localization of insurance products, which are usually heavily regulated by jurisdiction.

New projects joining the protocol will in turn create a ready market for other elements of the broader Etherisc ecosystem, whether they be oracle providers, KYC services, licensed insurance companies who legally ‘rent’ their licenses by allowing firms to piggyback on their own licences, relayers, data scientists with unique risk models to cover entirely and yet-uninsurable forms of risk, or claims adjusters qualified to serve as trusted authorities for traditional insurance. As the network grows, additional capital should be attracted by the opportunity to pledge funds to the risk pools functioning as reinsurance pools within the network, standing ready to insure the larger tail-risk scenarios that might require excess capital beyond the immediate capacity of a small insurance dApp.

A Platform for New Insurance Products

This design is intended to support the emergence of new insurance products while also enabling forms of insurance recognizable today. For example, insurance products for smart contracts or cryptoassets could emerge if entities like OpenZeppelin could step in as underwriters to assess smart contract vulnerability, cryptoasset holders stepped in with risk capital, and developers and cryptoasset users purchased policies. Etherisc imagines the protocol as a potential platform for anyone to monetize an ability to develop better risk models than those that exist today or more specialized risk models in niche markets, for example, rainfall in certain specific geographic areas.

The DIP Token

DIP tokens are used by all service providers throughout the system–any entity earning fees will be required to stake tokens in a proportion between 15-25% of their monthly income on the platform. Users are not required to use DIP to pay for insurance products or receive it as payouts. Given that premiums are collected on-chain and fee structures should be written into the insurance product smart contracts, this monthly income should be easily calculable, though the method of enforcement and potential punishment for noncollateralized activity is still being determined. If the requirement can be seamlessly enforced, DIPs value should track both overall adoption of the platform, as it is a ratio of realized revenue for all participants on the platform, and the on-chain fee structure of the participants.

For example, if $120 billion were purchased in insurance products, with an average total aggregate service provider fee of 10% resulting in $1 billion in collected monthly fees between Keepers, Relayers, and other service providers, then at least $150 million to $250 million worth of DIP would need to be staked. For casual comparison, in 2017, $12.6 billion was issued for catastrophe bonds in 2017; 2017 total global revenues for flight insurance were an estimated $19 billion. This suggests Etherisc will likely need to expand to more traditional forms of insurance to reach such numbers: homeowners in the US spent an estimated $91 billion in 2017.

The primary purpose of the staking mechanism appears to be incentive alignment: all operators on the platform benefit from its success. The staking mechanism also does serve a purpose with Oracles, because an Oracle’s stake can be forfeit if they don’t return data within a Keeper-defined window of time and distributed to a back-up Oracle, but similar punishment mechanisms for staked DIP for other service providers haven’t been defined. As such, the DIP token is not part of a cryptoeconomic incentive mechanism to prevent collusion, fraud, or other bad behavior, and while the team could implement such mechanisms in the future, they have built much of the system around off-chain incentive mechanisms (insurance regulatory regimes).

The token also has an inflationary rewards mechanism designed to incentivize the development of new products and additional platform features.  New tokens will be emitted to any entities bringing new risk onto the platform, using a method still being finalized. The intention is to incentivize the growth of insured risk existing on the platform, while also making the token more valuable to network participants–who must stake DIP in order to operate as a contributor–than to speculators.

2018 DIP Token Sale

The token sale successfully raised 8038 ETH, worth approximately $3.7 million. They sold 45,478,958 tokens, implying a network valuation of $81,400,000.

The sale involved a total of 300 million of the 1 billion total token supply via a Swiss-based sale requiring KYC and AML from all participants. The token price is $0.10 per token, with a hard cap of $30 million USD, and no minimum sale threshold. While not offering pre-sale discounts, token purchasers could access a 10% discount by agreeing to lock their tokens for a 12 month period.

The Etherisc Team

Ethrisc was founded by Christoph Mussenbrock, Stephan Karpischek, and Renat Khasanshyn in 2016. The cofounders have considerable experience in IT, startups, business strategy,and insurance markets that suggest a long-term interest in the field and the technical knowledge to push the Etherisc project forward. The project’s advisory board, including such well-known names as Ron Bernstein, Ralf Glabischnig, William Mougayar, Jake Brukhman, and Tobias Noack suggests a well-connected group that should not struggle to access the resources required to move their project forward.

They have also been operating in the industry for awhile. They were one of the winners of the global hackathon event hack.ether.camp in 2016, and the evolution of their ideas from then to now demonstrates a commitment to the vision of decentralized insurance. They have also already deployed MVP insurance products: members of their network recently used FlightDelay to insure a set of flights using smart contracts deployed on the Ethereum mainnet, and several members earned an ETH payout due to flight delays.

Etherisc and Decentralized Insurance Protocols: Opportunities and Challenges

In attempting to pioneer a range of blockchain-based insurance products Etherisc represents a genuinely Innovative attempt to expand blockchain-based products and services into virtually unexplored new fields. In doing so, the Etherisc team has several significant opportunities before it and has also taken on an extensive set of challenges, both technical and industry specific.

Opportunities

A live Ethereum mainnet MVP: The team has already built a flight insurance app covering a range of international airports around the world and carriers. Users can enter flight information, self-confirm they are actually booked on the flight, and see the required premium, payout amounts, and payout conditions. It accepts ETH, USD, EUR, and GBP (though ETH is disabled at the moment). An example policy for a flight from San Francisco to Chicago will accept a $20 premium and payout almost $300 if the flight is over 45 min late and almost 500 if it diverted or cancelled. Flight Delay Dapp has a license from Atlas Insurance PCC Limited, an insurance company based in Malta, that is the legal vehicle issuing policies. Their Dapp has issued ETH payouts in the past, when it covered several hundred developers flying to Shangai for Devcon 2. They are reportedly working on a range of additional products.

The demand for risk capital is high and growing. Catastrophe bonds, which are used to insure against catastrophic weather and other natural events, are both a source of real returns and a smart addition to a balanced portfolio, since catastrophic weather is not correlated to market performance. The tokenized reinsurance pools that Etherisc imagines underlie Keeper policies could be a financial vehicle for investors seeking tradeable exposure to risk of varying types.

In addition, niche insurance products have emerged in recent years, including flight insurance and third-party consumer electronics insurance, suggesting consumers are looking for insurance against risk too.

Designed around the modern insurance industry: The team is very well versed in smart contracts, the insurance industry, and blockchain technology generally. They have designed this smart contract based protocol around current restrictions and actors in the insurance industry today. This has enabled them to launch compliant insurance products on a public blockchain, while opening a path for the insurance industry to migrate to Etherisc with minimal discomfort.

Challenges

Traditional well known challenges in offering insurance products, such as information asymmetry, verification of triggering events, and establishing and proving loss, won’t be discussed here. Etherisc relies on many regulatory regimes that structure current insurance markets. This design choice has implications, which will be discussed below, but it also means Etherisc is not alone in facing them.

Cryptocurrency Volatility: There is also an obvious concern with using a volatile currency like ETH as premiums, since all parties take exposure to both risk and ETH prices, but this is a challenge many projects in the industry face and nothing in the Etherisc protocol prohibits the integration of a stablecoin.

Heavy reliance on off-chain incentives and coordination. The balancing set of incentives pushes some of the intractable complexity of insurance upward to off-chain operations rather than directly addresses them. The list of such incentives includes the following:

  • Users won’t commit insurance fraud (for example, by claiming they planted a field when they actually did not) because fraud is punishable by law. There is no on-chain mechanism for addressing this, and it is presumably on Keepers to define the evidence they require, as currently exists in the world of traditional insurance. Flightrisk can rely on readily available APIs to prove both flight delay and other methods to verify purchases, but not all insurance applications will enjoy this kind of digital infrastructure.
  • Keepers and Relayers won’t collude to offer fraudulent or exploitative insurance policies, because such activity would be punishable by law. Relayers also presumably have reputations to maintain, so they will help keep badly designed policies away from their website.
  • Keeper interests (in not conducting payouts) and user interests (in maximizing payouts) in cases where payouts are based on estimated loss should occur
  • Automated on-chain payouts would presumably be protected from compromised or colluding oracles through an off-chain oversight process. For example, any sensor readings that would result in massive payouts would likely need to be verified several times before issuing payouts, preserving some of the friction (administrative costs) that traditional insurance deals with.

The possible result could be that little insurance innovation takes place. For example, a broad-based Relayer realistically isn’t in a position to evaluate independent Keeper risk models and likely doesn’t want to sell bad policies to consumers. As such, they will look for established license holders they recognize. Licensed insurance providers likely won’t back insurance products they themselves don’t understand and that don’t resemble the economics they already face, including administration costs and shareholder pressures. The result could be a slow migration of existing insurance products to an Etherisc-based accounting system, with a range of niche insurance products that don’t rely on third-party Relayers.

Etherisc also does not address how all these parties will discover each other and collaborate. Rather, the assumption is that coordination happens off-chain, and on-chain contracts represent the resulting insurance product and professional relationships. Building mechanisms for on-chain coordination or on-chain incentive mechanisms for good behavior (in addition to that described for oracles) would bolster the argument that this is a protocol for decentralized insurance rather than a smart contract accounting system for centralized insurance products.

Insurance against completely unknown risk, such as smart contract security, is very difficult. There are few entities qualified today to both estimate AND value the likelihood of smart contract failure. In order for those products to come to market, they need risk capital, but risk capital presumably would want some confidence in the risk models being used to govern premiums, payouts, and their own return. One might imagine specialized risk capital providers (like existing crypto holders who want to see Etherisc succeed) stepping in to kickstart the process, but this dilemma stands in the way of Etherisc being a protocol for entirely new insurance products to reach scale. In some cases, it also presents several obvious attack vectors (developers hack their own smart contracts) that would need to be mitigated.

While none of these challenges are necessarily fatal ones, and solutions are likely to emerge over time, they illustrate the range of issues the protocol will need to confront as it looks to scale both the size and range of supported products.

A Broader Theme: What is Decentralization?

On a larger conceptual level, there is a core tension that runs throughout the Etherisc endeavor that is indicative of tensions within the overall industry and facing many blockchain entrepreneurs. This tension is between the desire to create a self-contained distributed, blockchain-based ecosystem, as many blockchain-based projects seek to do, and the inability to escape the off-chain world of state authority, regulatory environments, and associated penal codes. For example, an adjustor’s off-chain judgement will likely be required to ascertain whether a field was planted, or planted in a satisfactory manner; threats of criminal penalties for insurance fraud are a critical incentive mechanism for Keepers, Relayers, and policy purchasers, much like they are today.

In some ways, this might represent a realistic assessment of how to innovate within an industry as established and regulated as insurance. It is not difficult to imagine the scale of risk in promoting a decentralized insurance protocol designed to disrupt the insurance industry by offering completely unlicensed and unregulated insurance products. It’s simply not clear that smart contracts solve the array of well known challenges bedeviling insurance and which have given rise to the range of restrictions that exist today. Etherisc has opted for a more realistic path of adoption–one that allows them to compliantly develop a range of MVP insurance products like FlightDelay while exploring partnerships with existing insurance providers that involve readily recognizable roles. The insurance products currently in development are testament to creating an ecosystem structured around how insurance operates today.

In other ways, it also represents a design heavily based on the industry as it exists today rather than how it could or ought to exist in a more decentralized form. The approach still relies on off-chain validation and cases of manual claims adjustment, proof of off-chain re-insurance, and enforcement to protect against collusion between entities like claims adjusters and insurance product providers. One risk of such a design is that it could simply become a more efficient back-office settlement system within the existing industry, while supporting a range of small niche products not touched by existing insurance providers and not easily discovered by consumers. This could undoubtedly lower some administrative and transaction costs, but without changing the way insurance incumbents themselves operate, would also not address the most complex aspects of the industry. This scenario would represent a much slower path to insurance innovation, though the benefits of using smart-contract based accounting would be real.

This is a tradeoff that faces many projects today that seek to apply blockchain technology to well known problems but aren’t interested in launching systems that enable wildly illegal or likely non-compliant practices. It’s unclear how this tension will play out, but many token-based projects have faced similar tradeoffs. The industry would undoubtedly be different today in many ways if Satoshi had designed Bitcoin specifically around existing financial regulations, potentially resulting in a cryptocurrency less prone to use for things like Crypto-locker attacks and Silk Road purchases but also not a decentralized currency no one, especially not financial incumbents, controls.

Watching Etherisc navigate this tension as it grows should yield insights for the entire industry.

Conclusion: The Future of Blockchain-based Insurance

Etherisc’s development of a protocol intended to support the pioneering of blockchain-based insurance, with a particular focus on the emergence of products for niche markets that are often underserved by existing insurance companies, represents an intriguing addition to the range of industries where blockchain-based products offer the promise of disrupting established industries. Potential successes for the protocol and dApps building upon it promise to create new opportunities for consumers and business, while eliminating middlemen and creating a more responsive market better able to serve individual needs. Etherisc is clearly poised to find an enthusiastic reception, whether from critics of the insurance industry ready to embrace efforts at innovative, cost-effective, and consumer friendly new business models, while participants in the sector are likely to welcome Etherisc’s promise of opening a variety of new markets.

A supportive market context is a useful background for the appearance of the Etherisc protocol, but the challenges facing efforts to develop a decentralized insurance protocol are numerous and substantial. Observing the development of the Etherisc protocol and dApps built upon it, and more broadly of different attempts to develop blockchain-based insurance protocols, promises to be a significant field going forward.

The positive impacts of the successful development of Etherisc’s ecosystem would, in a similar manner to claims made about stablecoins, become a crucial piece of infrastructure underpinning the continuing expansion of the crypto community. The potential for individuals to insure their assets and risk exposure in cryptocurrency, and for traditional investment capital to find opportunities also in crypto, would be particularly meaningful innovations should Etherisc develop and successful scale across a range of markets.

Observing the progress of both individual projects built upon the platform as well as the platform itself will be a rewarding activity going forward, since Etherisc is tackling a number of daunting challenges that will not easily be addressed. While the platform’s structure is a thoughtful one that suggests it is perhaps likely to survive even if the initial projects built upon it struggle to develop significant user bases, it may be some time before the collective activity upon the platform translates into a flourishing token economy.