As blockchain technology continues to percolate financial services, Kyle Owens and Matthew Queen provide an overview of how smart contracts could be implemented in the insurance industry — and what implications this may hold for captives
It is not an exaggeration to say that, by now, the majority of people are at least somewhat aware of cryptocurrency and the blockchain. Most online news attention is devoted to cryptocurrency centres around the success and failures of speculators making and losing their fortunes overnight.
Blockchain technology will likely have a lasting impact on finance, accounting and insurance more than the fast-rising popularity and subsequent fall of the hundreds of cryptocurrencies that currently command million- to billion-dollar market capitalisation valuations.
What is the blockchain?
A blockchain is a distributed ledger that is completely open to anyone. It has an interesting property — once some data has been recorded inside the blockchain, it becomes very difficult to change it. Each block contains:
- Data: the data that is stored within the block depends on the type of blockchain. For example, Bitcoin blockchain stores the details about a transaction, such as the sender, the receiver, and the amount of coins
- The “hash” of the block itself: you can compare a hash to a fingerprint, as it uniquely identifies a block and all of its contents. Once a block is created, its hash is calculated by computers as a string of seemingly random numbers. Blockchains are programmed so that changing something inside the block will cause the hash to change
- The “hash” of the previous block: this effectively creates a chain of blocks, and it is this technique which makes a blockchain so secure
But using hashes is not enough to prevent tampering. Current computers are fast and can compute hundreds of thousands of hashes per second. You could tamper with a block and recalculate the hashes of all other blocks to make your blockchain valid again. To mitigate, blockchains have “proof-of-work”, along with other methods of verification.
The security of a blockchain comes from its creative use of hashing and the proof-of-work mechanism, but there is one more way blockchains secure themselves: being distributed.
Instead of using a central entity to manage the chain, such as a central bank, blockchains use a peer-to-peer network which everyone is allowed to join. When someone joins this network, they get a full copy of the blockchain. Each individual copy of the ledger is a “node”. A blockchain node can use this to verify that everything is still in order.
When someone creates a new block, it is sent to everyone on the network. Each node then verifies the block to make sure that it has not been tampered with — if everything checks out, each node adds this block to their blockchain. All the nodes in this network create consensus, and blocks that are tampered with will be rejected by other nodes in the network. To successfully tamper with a blockchain, a bad actor would need to tamper with all the blocks on the chain, redo the verification for each block, and take control of more than 50 per cent of the peer-to-peer network. Only then would the tampered block become accepted by everyone else. It is possible for governments to take a digital offensive and disrupt the chain. A captive insurance company can provide protection by providing insurance over the blockchain.
What are smart contracts?
In simple terms, it is the use of the distributed ledger to store contracts. They are just like contracts in the real world — the only difference is that they are completely digital. In fact, a smart contract is actually a small computer programme that is stored inside of a blockchain.
But why should we trust smart contracts? Stored on a blockchain, tampering with smart contracts becomes almost impossible as they have two key properties:
- They are immutable: once a smart contract is created, it can never be changed again. No-one can tamper with the code of the contracts
- They are distributed: the output of the contract is validated by everyone on the network. A single person cannot force the contract to release the funds, because other individuals on the network will spot this attempt and mark it as invalid
Smart contracts stand to replace many middlemen historically relied upon to facilitate a transaction. Using any number of cryptocurrencies, the time for transaction finality is reduced. Instead of transactions taking days to process through a bank, transactions are instead finalised nearly immediately.
An infinite number of “wallets” can be created, each within seconds, acting as de facto bank accounts without the inclusion of any third-party. Each wallet can send funds directly to any other wallet. The blockchain itself would include that transaction in one of its “blocks”, the transaction would be processed by many computers using proof-of-work, and then the transaction would be settled with no bank accounts and no third-party services.
The implications to traditional banking are enormous. The peer-to-peer nature of cryptocurrency divides it from traditional currencies defined by a sovereign government. Crypto needs no such designation, and a transaction involving crypto can occur without a central bank.
Implications for insurance
The main issue here for insurance is fraud. It is all well and good that contracts could be programmed to gather and release money without a third-party, but don’t we need inspectors to take a physical look at the damages and assess whether or not a claim being made is fraudulent? Not for everything. The best use case of this might be life insurance, where there are two possibilities: either an individual is dead or they are alive.
The only problem to solve is determining a mechanism to verify this fact.
Perhaps the contract is programmed to monitor a registry provided by a local government of all deaths and births within a given jurisdiction. Once an individual matching a given set of characteristics (name, age, birthday, and other identifiers) is found by the contract on the registry, the money is automatically released to the insured because everything relating to the contract that needs to happen would be programmed to automatically take place.
In this scenario, no one needs to fill out paperwork or call a representative unless there is an issue, although the better that such contracts become at performing correctly, the less this might happen.
The issue with other kinds of insurance is that damage is a matter of degree, not an either-or proposition. The question is the extent of the damage, not whether there is any damage. This becomes difficult in terms of automating the processes that could solve this problem.
There are various solutions that could in theory mitigate this issue, however. The insurance company could delegate the assessment of the damage to a specific chain or network of auto body repair shops. If the shop certifies that certain damages have been incurred, the smart contract will pay out different amounts depending on that information.
Alternatively, it could examine rainfall data provided by some third-party for different parts of the country; if rainfall is below a specified amount for an extended period of time in a particular area, individuals owning farming land would collect on that insurance without the need for inspectors to physically view the area. These are just hypotheticals, but the theme is the same: delegate authority to an unbiased smart contract and have it refer to a trustworthy data source in order to determine when and how the funds in question are to be distributed.
When it comes to captive insurance, a similar landscape exists. If smart contracts were created for every single risk that the captive is insuring, and if there was a verifiable method of automatically discerning when the conditions of the contract have been met, large amounts of paperwork and work hours could be reduced to a fraction of their current state. However, as many captive insurers are not highly staffed, the cost savings would likely be less than that of a non-captive insurer.
In a utopian future, governmental entities in charge of setting the rules for the captive insurers, both state and federal, would allow captive insurance companies to be run entirely by smart contracts without the need for any physical participants whatsoever to oversee it.
There would be a standard set of smart contracts deemed reputable by the government which would be available for use by would-be captive insurance companies. The insured entities would still send premiums to the captive insurance company wallet on a given blockchain. This wallet would be separate from the insured, and the insured would not have access to the wallet nor be able to tamper with it in any way. Once certain conditions are determined to have been met by the smart contract, it then releases the funds back to the insured.
But we are far from such a future — or not far at all, depending on who you believe. But consider the fact that we still do not have fully autonomous vehicles in which no driver is required whatsoever. The issue is that there are too many “edge cases”. In autonomous driving, this refers to low-probability occurrences happening on the road that the computer has never had to react to before in that specific context, and then makes an error doing an automated procedure resulting in a crash.
In finance and accounting, it is items that involve subjectivity or judgement, for example, an item’s market value, or interpretation of a certain set of laws. However, it is not hard to see how these things, that are highly numerical in nature, could more easily be given up to smart contracts. For example, market values are easily found or estimated by computer programmes. The key is that enough data exists for the automated process to be accurate in its assessment. As the technological age rolls on, and because more data is collected now more than at any other time in human history, the ability of smart contracts to properly operate will increase.
Looking ahead
The main effects that blockchain and its derivatives will have on insurance, and captive insurance specifically, will mainly be to give greater efficiency and thus lower cost to perform certain types of insurance transactions. Certain insurance companies may even produce their own cryptocurrencies as a payment system on their blockchain. With smart contracts, loss adjusting expenses can be significantly reduced.
The effect of this would be to finalise transactions much faster between participants on the network. It would also have some effects not specifically confined to insurance, such as incentivising individuals and organisations to hold and invest in their cryptocurrency by mechanisms such as “burning” tokens. It could also impact firms as they distribute a percentage of fees collected from transactions on the blockchain to entities that hold the token as a sort of annual percentage rate on their holdings, and essentially capture funds this way and keep them within the blockchain owner’s ecosystem.
Finally, the cryptocurrency could be utilised as an investment in the hope that as the blockchain itself increases in value, so too will the tokens on that blockchain — but these effects are tangential to actual insurance.
