Currently showing: Funding longer lives > Social contract


31 Dec 16 10:01

During the last year, I've been working on Blockchain at Swiss Re. In my work on the technology, my interest and focus has been on investigating practical ways to make it useful and create business benefits for insurance and reinsurance. You may have heard about the B3i (Blockchain insurance industry initiative), which we started this year to look at how we can roll out Blockchain industry-wide. I think it's safe to say that our industry has now truly discovered Blockchain. But you're probably wondering, "Is Blockchain just a hype?" So today, I will try to answer that simple question: "Will it be a game changer for our industry?" Here are 10 reasons why it may.

# 1: Blockchain is easy like lego
The simplest first: the beauty of building something block by block. Our children do it with lego or dominos. In insurance we often work with multiple versions of contracts. They form the basis to collect premium payments and reimburse claims after loss events. We use account statements to exchange information about what happened. These are our "building blocks". Each block only makes sense if connected to another. When you have built a giant lego castle, you cannot take out any block in the middle of it and try to replace it without being noticed. Blockchain works in a similar way and that is a good thing, especially in insurance, which relies on trust and truth around events, properties or people. An immutable mechanism to trace and manage authenticity and versions of data can be a breakthrough, as it will eliminate inefficiencies, reduce fraud, mitigate operational risk and strengthen contract certainty.

# 2: Latency is legacy
Executing an  insurance contract or settling a claim can take time, a lot of time, with limited overview of a payment’s status until it has fully traversed a cross-border chain. The so called "batch processing" leaves parties unaware of the settlement while the beneficiary receives the credit only after a delay. Reducing this latency, not only of payments but the entire upstream contract and settlement cycle, mitigates risk, improves liquidity management and reliability of financial reporting. Rather than being in the dark, it will be much clearer when transactions settle. This can be a "what's app" moment for insurance, enabled by Blockchain's smart contract handling on a distritubed ledger.

# 3: Omnipresent means sharing
Underwriting or claiming insurance often involves agreeing on different versions of the truth, as it relates to assumptions, scope, cover, facts or circumstances. Is the claimant really the owner, does the property really exist, what is the industry consensus on the large loss event? Shared ledgers with shared logic, covering the same events and actions recorded at the point of occurrence, provide an instantaneous and shared view of status. Sharing data in a trusted central place makes the management of authenticity and identity much easier.

# 4: Consensus can be efficient
Judgement or plausibility checks can be required for contractual parties to reach agreement. If such consensus is found between contractual partners, without relying on a central authority or intermediary, things can really get efficient. This does not mean a fully machine driven process. Human interaction will still be key (e.g. for underwriting and claims adjustment sanity checking or auditing) and a blockchain signature authority mechanism can process such human approval steps. The overall process of reaching consensus will become leaner. Insurers then use their reserves primarily to compensate for losses, rather than to fund disproportionate administrative, transaction or legal fees. It may not be intuitive but law and computer science are quite natural partners, or with a quote from Nick Szabo, the man the man widely credited with inventing the smart contract concept itself: "Traditional law is manual, local and often uncertain. Blockchain smart contracts are automated, global and predicable in their operations."

# 5: Kata – when we collaborate, we make things better
This lean mindset advocated by Toyota relentlessly pursues continuous improvement in cost, quality and service. Processes typically improve when people collaborate. That is why you see teams huddling together on lean managed factory floors. Insurance dearly needs this, to address the cash and paper flow trapped across its value chain. The insurance industry now has a unique, common platform to jointly run such improvements and make insurance more accessible and affordable to all.

# 6: Cash is king
Time value of money is key to insurance, surely in times of low interest rates with high opportunity cost. It further increases the pressure to improve the cash conversion cycle and optimize asset-liability management. This requires reliable cash forecasts from re/insurance operations. The engine for such forecasts are quarterly settlement processes, which can take many months to settle, leading to credit risk, unreliable forecasts and inefficient usage of liquidity. A distributed ledger, combined with latest payments message standards can bring a break-through in the payments and cash management cycle. Why can we transmit a private video message across the globe in seconds, but need several days for a payment? This is the "last mile" that needs to be solved to address the latency issue described above. There are examples already that prove this use case, even if they may only leverage particular parts of blockchain technology, such as cryptography and instantaneous settlement, that add most value to payments.

# 7: Holistic vision rather than proprietary
Insurance participants handle massive amounts of data via various platforms. The industry has already developed partial e-admin solutions, taxonomies or exchanges, but with adoption limited to certain portfolios, segments or markets. The industry needs to depart from its proprietary legacy towards a holistic solution, much like what SWIFT became for banking and preferably even more courageous, more efficient and open source.

#8: Actors not just altruists
Is it a too simple and altruistic view that the industry actors can come together for something that may not bring competitive advantage to any single participant? Insurers, brokers and reinsurers will have to choose their act in this play. Standards exist already that can be leveraged. Unlike banking however, the insurance industry does not have an industry-wide network to operate messages with this standard. Blockchain could fill this gap. It will reduce time for shuffling cash and paper around, and free up time to manage risk and portfolios.

# 9: Immutable means "first time right" thinking
In large-scale insurance contracts, parties self-declare how their portfolios have developed. They reveal facts about business development, exposures, losses or reserves, yet without running competitive risk. It is good to apply a "first time right" mindset to such reporting, with a clear audit trail of every change to the initial declaration. This will eliminate much reprocessing and reconciliation work that adds no value. A distributed platform for estimates where updates are traceable, auditable and most of all immutable is new and powerful for the insurance industry and can dramatically improve its business processes.

# 10: Neutral
Insurance implies periodically reporting mutual obligations. Parties run these data through the terms and conditions of the contract that binds them. The outcome should be the same, only with reverse signs. Yet a lot of time is spent performing separate calculations, hoping to arrive at the same result. When multiple parties share the same contract logic, the benefit of executing the contract in a shared, neutral ledger is obvious.

These are not yet full fledged solutions but ten knots which, if unraveled, will make up a game changer. My recent experience tells me that, if the industry collaborates and combines its business and technical knowledge, there has never been a greater window of opportunity to make this happen. As Nick Szabo once told me: "I don’t think a computer can understand most real-world conditions, exclusions, or the general variety of damages, but anything in insurance that is automated and not intensive in terms of computation or data storage can be done on the blockchain where there would be benefit from the added security and reliability."

 If you think this makes sense, come back and read my more specific findings on these topics in the next series or write me a comment below. I look forward to a discussion!


Category: Funding longer lives: Social contract, Climate/natural disasters: Resilience, Other

Location: Zürich, Switzerland


14 Comments

Alain Beland - 5 Jan 2017, 2:19 p.m.

Hi Paul, thanks for sharing. I read this and I had also read the annoucement about the B3i initiative. I am keen to understand more. However, I still don't fully get the basics, i.e. what is the actual technology breakthrough with Blochchain. I am sure you have a good reference to share!

Paul Meeusen - 7 Jan 2017, 5:50 p.m.

Hi Alain, at the risk of incompleteness, in a nutshell, blockchain is a technology that permanently records transactions in a way that cannot be later erased but can only be sequentially updated. The printing press was excellent to distribute information. Then came the internet, which was even better and faster, via the TCP/IP data packaging, but it is still copying information from one place to another. This is not a good idea when dealing with items of value, like money, unless you are the central bank. So the next protocol layer that blockchain brings, is one that allows not just to transmit information but to transfer anything of value (identity, will, title, almost any type of financial instrument). Such decentralised system of record or a distributed ledger can efficiently enforce multilateral consensus on every transaction recorded.

Again at the risk of selecting among the wealth of information around, nonetheless a few recommended references:1) "Revolution: How the Technology Behind Bitcoin Is Changing Money, Business, and the World " by Alex Tapscott who will also speak at a a conference organised by Swiss Re later this month: http://cgd.swissre.com/events/Blockchain_insurers_are_building_up_trust.html#tab_3;
2) Some publications will also be issued around this event. One publication that I would recommend to particularly watch out for is of the author Kamesh Raghavendra from The Hive, who explains key features such as immutability, transparency with privacy, consensus building and smart contracts and how this specifically can be applied in the areas of risk management, claims and underwriting. His suggestion of using smart contracts and digitally executing legal covenants with templates for specific arrangements like quota-sharing (QS) or excess-of-loss (XL), is exactly what we have done in the retrocession prototype at Swiss Re.
3) Finally, I find anything of Nick Szabo always good read, being a PhD in both computer science and law, and quite "hype resistent" since he started publishing about smart contracts years before the blockchain "hype". Also - my personal view - I like him because (a bit like Elon Musk) he is not your typical slick super presenter, so you need to keep focused on substance over form, here is a good recent example:https://www.youtube.com/watch?v=v_-mxyN4pcY
Hope this helps.

More practically, in Swiss Re, the Digital strategy team of Elizabeth Wesson and Jags Rao, and the Blockchain Center of Competence of Andreas Fink are also excellent resources to consult.

Alain Beland - 15 Jan 2017, 8:57 a.m.

Great Paul, thanks for the information.

Andreas Lindemuth - 5 Jan 2017, 3:36 p.m.

Thanks for the article, Paul.
Actually, the very same question mark that Alain brought up popped up in my mind: what is the actual technology that can make all this happen, and why did it not happen long before?
Also, what I would really appreciate is another blog from you called "10 reasons why Blockchain may have vanished into thin air in 5 years from now". It is very easy to find out why there is a hype around Blockchain; you just google, and you will be drowned in material. What is much harder to find is literature that gives you a balanced and realistic view of the potential behind a new technology. So I think an article that is somewhat of a counterstatement to your theses above would be of huge value to many of us.

Paul Meeusen - 7 Jan 2017, 8:16 p.m.

Thanks for the feedback Andreas. Why it did not happen before, you can hopefully take from my previous response to Alain. I tend to react to innovation, neither by blind submission, nor by complete denial, but rather with curiosity and experimentation. A year ago, a good colleague at Swiss Re, Jags Rao, introduced me to blockchain. I was first sceptic myself, left with many questions and also curious. It struck me that the proclaimed capabilities were quite relevant to issues that I had seen for years in insurance finance processes. In particular I thought about the cash settlement of large transactions, such as IGR contracts. After consulting with contracts and IGR specialists, such as Rory Unsworth, Ariella Pfenniger and Richard Phipps, we focused on building a prototype for a QS (quota share) and a XoL (excess of loss) contracts. This was completed within a few months. With help of coders from places such as Latvia and Ukraine, our Swiss Re IT specialists, Philippe Tölle and Sebastian Müller, worked side-by-side with them, in a very complementary and open-minded spirit. The development was much faster than what I was used to in other financial system developments. Key was that we did not just focus on technology, but on addressing business issues such as reducing cycle time, transparent audit trails and sign-off and fast settlement. Our view was still balanced, with many open issues such as scalability, privacy, type of data standards (e.g. Acord) to use, data integration etc. So we applied another basic principle: listen to the customer. We shared our prototype with clients and peers. A few months later, today, 14 peers have joined the B3i initiative to further develop this prototype and test if it can function as an inter-company, industry-grade platform. This is still an experiment and we go into it open-minded but with a strong belief that this is adding more value than just producing literature.

Andreas Lindemuth - 9 Jan 2017, 12:08 p.m.

Hi Paul. Thanks for all the explanations; the resources you gave provide some valuable information.
I completely see the potential of decentralised ledger + smart contract technology in the banking industry. There could be significant efficiency gains for transactions ranging from simple money transfers to complex derivatives (although even here we have to keep in mind that the 2% transaction fee we pay to a bank is not necessarily needed to cover the frictional costs caused by the transaction execution; the fee might cross subsidise other activities of the bank, which blockchain technology cannot automate, such as the structuring and markerting of a financial derivative).

However, I still struggle to see the same potential for the insurance industry. The fundamental promise of a (re)insurance agreement is that "I will pay for your future losses under certain conditions". Now these losses and also those conditions are typically something that takes place in the real world, not in the virtual world. No matter how smart a contract is, the losses and conditions will still have to be "manually" verfied in the real world, and this is what causes major admin expenses in the insurance industry.
Also, ambiguity is embedded in the very nature of the promise. What exactly is a "loss", when exactly are the "conditions" met?
A smart contract could handle this ambiguity in two ways: either the smart contract code only reflects the unambiguous elements of the agreement (like applying deductibles, cover limitations, etc) and leaves the ambiguous parts (was WTC one or two events?) to decisions taken in the real world ... but this would not eliminate any of the inefficiencies.
The other way is that the smart contract code anticipates all possible scenarios that may impact the agreement, and holds a pre-programmed solution for each scenario. This would certainly make the contract execution more efficient, but the administrative burden would simply be shifted to the pre-signing phase, when all those scenarios need to be worked out (I do not want to be the one doing this job :-) )
What are your thoughts on this?

Doug Kraft - 5 Jan 2017, 10:33 p.m.

I don't mean to pour more on, but I had similar questions to the 2 above. If Blockchain results in everyone in the world using the same data standards and definitions, that alone could be a huge breakthrough. How much time and effort are spent translating data - figuring out what "premium" means from Client A vs Client B?

Maybe even more importantly, how often do we make bad decisions/assumptions off of something that's been miss-mapped or missunderstood b/c of this? A sort of "universal admin system" would be gamechanging. But, would this really be the silver bullet against fraud, timeliness in approval/declines, judgment of exactly how much damage was caused by a given event? This is where it would seem like "hype" to me as these are human decisions.

But, just "going digital" in and of itself is old news - eg, most of us probably live most of our lives without actual cash. Swiss Re pays me by sending numbers to my bank. I buy most of my food and goods with my credit card. Then, the bank automatically just takes some of the numbers out of my account... my retirement fund is even more ethereal with numbers and graphs just moving up and down daily. What someone used to be worth "on paper" is now what they're worth on someone's server or "the Cloud". That is, most of the world already exists/transacts digitally without "Blockchain" (with the exception of cabs in Hong Kong...), so what's changing?

Paul Meeusen - 7 Jan 2017, 10:17 p.m.

Doug, as you may have read in Nick Szabo's article that I referenced, lawyers should not worry, exercising judgement will still be needed, where other tasks can be automated by smart contracts. Much of what we have digitised however, is still just a digital copy of old processes. Take cross-border banking which can still take several days or FX trades that your bank or credit card easily charges you 2.5% for. That can be quite relevant in, say, high value claims payments.

Yan Choi - 6 Jan 2017, 4:19 a.m.

Thanks Paul... now I (from L&H Claims and also some Underwriting) understand more about blockchain.

My read is that it can increase trust of what you see in documents.

How will this impact below fundamental challenges of insurance?

(1) insurers knows less (health and etc) information than what is needed to make insurance work. I guess yes - blockchain will impact both undewriting and claims. It's hard to see when and how... depends on whether customers PROVIDE information. For third-parties information source (e.g. doctors for L&H insurance, auto-repairers for motor insurance), Data Protection law and culture come into play.

Another thought is what is "truth"? What if the blockchain documents are different from the physical fact from the start? Seems blockchain is robust against inconsistencies but if the whole lot is consistently untrue then it could be overriden by the physical truth or self-declaration of such. (e.g. I actually had no pain in my stomach in Apr 2013)

(2) customers not sure if in the future the insurer will be around to pay as promised. I guess no - blockchain will not impact. Insurers like all companies may go broke. We know it and the customers know it. Increase trust in information doesn't help a lot to increase trust in solvency.

Hope the above contributes to the discussion. :-)

Paul Meeusen - 7 Jan 2017, 4:14 p.m.

Thanks Yan. I like how you start with the business challenges, than see if new technology can help, rather than the other way around. With (1) you rightly point to the information rich society we live in, making it challenging to distinguish noise from signal. As insurers, and surely actuaries, I assume we prefer less information distortion. The more personal and subjective such data however, challenges arise as what you describe, which probably explains that first successful applications have been rather in identity management of objects (like diamonds, see the everledger.io example I refer to in the article) or weather events (parametric data transmitted to smart contracts). On (2), I agree that technology is unlikely to solve the solvency equation, with perhaps exception of smart contracts auto-reporting certain triggers, such as collateral, capital adequacy or other rations, applications that have been tried out on blockchain.

Theodor Bachmann - 12 Jan 2017, 8:43 a.m.

Thanks, Paul, for a very comprehensive view on the opportunities of Blockchain in Insurance.

From a P&C Claims and Accounting perspective, this will - if widely adopted - help us materially in two ways:
First, the underlying standards for insurance/reinsurance claims movement, technical accounting and financial accounting messages are in place since 2008 and used constantly (1.2m messages exchanged in 2015 among Ruschlikon participants). Blockchain will make it easier for new participants to use these industry standards and allow companies to harvest some benefits that otherwise need to be done in their (often comples) system environment (one-time data entry, audit trail).
Secondly, it points more to a peer-to-peer future than a hub solution like SWIFT. Given the concerns of international insurance/reinsurance companies about data security, sensitive information and control of data , this will - in my view - make adoption easier.

Steve Hales - 16 Jan 2017, 8:49 a.m.

Interesting discussion. My feeling is that it is still early days for us in the Insurance industry to really understand what could be the full effect of blockchain. The great thing is that we are actively investigating it via B3i, rather than waiting to be disrupted.

Arno Greter - 19 Jan 2017, 8:29 a.m.

Interesting read and discussion. Thanks for sharing your thoughts, Paul.

Christian Richter - 3 Mar 2017, 3:46 p.m.

Thanks for sharing your thoughts, Paul. Very comprehensive list of reasons in my view.

I personally would emphasize two points here:
a) the need to establish standard for messaging, transactions as well as contracts and claims statements in the industry -- this is what drove part of the efficiency play in retail and commercial banking (even of one could argue it increased the speed towards commoditization of banking services but the regulators contributed their part to this too). The insurance industry has some standards but the are only partially adopted which is what still keeps intermediaries and they marketplace platforms alive; sometimes with no real value-add
b) the need to establish a consortium that defines standards, drives out efficiencies and promotes innovation based on DLT that is solely born by industry players themselves and follows the open source ideas at its best; given what some banks already started (i.e. JP Morgan with EEA etc.) this is the most straight-forward and collaborative way to drive adoption in the industry

Im am thrilled to see B3i taking the initial steps towards the industry vision you have outlined in your article!


If you would like to leave a comment, please, log in.