Periklis Thivaios explains how smart contracts work, their potential benefits, and what we should be cautious about.

We read a lot on the undoubtedly huge potential that smart contracts on distributed ledger architectures will have in the financial services industry (banking and insurance). Alas, before excitement (hype?) takes over and technological determinism drives decision making, we have to raise caution in the use of smart contracts! It sounds like common sense, I have to admit, but common sense is often not that common in an environment dominated by the exuberance surrounding blockchain!

What are smart contracts on blockchain?

Smart contracts represent a condition-based, self-executable layer of code sitting on a blockchain infrastructure such as Ethereum. When a specific trigger is met (which could be anything that can be codified digitally, from time to weather conditions), the smart contract executes itself and, in a payment configuration, disburses the payment to the appropriate recipient.

The potential benefits are numerous. Having intelligence built in blockchain-based payment methods (which aren’t necessarily cryptocurrencies, but that’s another discussion) is a great development from existing ‘dumb’ currencies: payments can be linked to trigger events, bilateral agreements can be incorporated in the payout code, and so on. A great (albeit misunderstood) example are initial coin offering (ICO) tokens, where the payout of an ICO investment can be codified on an immutable platform, removing the uncertainty pertaining to the execution of the investment payment (but not necessarily the value or the timing of the payment itself).

Are smart contracts something totally new?

Obviously, smart contracts on a distributed ledger architecture is something new, yet the embedding of automated execution logic in code is not, and history can teach us from a number of examples where such automation isn’t desirable, albeit feasible.

My favourite example is the automated collateral posting system of Bear Stearns about 10 years ago, which would calculate/confirm the margin requirement and automatically post the required collateral in real-time. The automated logic, even though not built on a ‘fancy’ blockchain platform, was one of the causes of Bear Stearn’s folding when the bank-sponsored funds (particularly the High Grade Structured Credit Strategies Fund) started sinking and the bank’s liquidity haemorrhaged.

In other words, the automation of payments (on blockchain or not) has been available for a while. Yet, even though automation solves some problems, it creates other ones that may be more important to control. Practically, the reason for having t+1 posting of collateral and derivative settlements isn’t due to the absence of supporting technology, but because it makes good business sense to allow for ‘uncool’ humans to intervene and control the disbursement of funds (other than the interest margin made on the funds).

Automation is cool, but also risky

Imagine the implementation of insurance claim payouts on smart contracts. That’s pretty feasible to different degrees (on blockchain or not) and definitely appealing. The process is streamlined, payment speed is increased with customer satisfaction alike, and transaction costs potentially minimised.

Now imagine a catastrophe that results in the automatic payout of the insurer’s available funds. In my humble opinion, the risk of insolvency is greater than the risk of customer frustration. The ability to delay the payout and protect the insurer’s capital position is at least as valuable as the ability to be considered a leading blockchain implementer.

Similarly, the automatic execution of an agreement (whether a derivative or reinsurance contract) requires the availability of reserves for the payment. Ring-fencing such funds (in an electronic wallet or old-school escrow account) may not be the best use of a firm’s capital. The smart contract will execute, but if the bank account is empty, the outcome will be nix nonetheless.

Lastly, don’t forget that it’s sometimes preferable for our firm to default on its obligations to another one, rather than the other way around.

Automate because it makes sense, not because it’s possible

Smart contracts on blockchain or automation in general can be hugely valuable, but the senseless implementation of automatic rules on technological systems, simply because such rules are made easy by smart contracts or other technologies, may not always be optimal. Sometimes, it’s simply better to be dumb.

READ NEXT: Smart contracts in regtech?

Image by Alexey Godzenko,

About the author

Periklis Thivaios

Periklis Thivaios is a partner at True North Partners. He has been involved in the launch, management and financing of startups in financial services, retail and hospitality. He also has extensive expertise in financial analytics, innovation and entrepreneurship, and international risk and regulation. Periklis is a frequent speaker at international banking, insurance and entrepreneurship conferences, a writer and a consultant.

1 Comment

  • Most of the parallels you drew and as you called out alongside have been / have existing/existed pre-blockchain era of automation. The question should be is smart contract adding new risks or aggravating the existing risks in the process for which there is either a solution yet to be implemented, or there is no solution by virtue of the design.

Leave a Comment