In 2008, a paper titled Bitcoin: A Peer-to-Peer Electronic Cash System was published under the pseudonym “Satoshi Nakamoto”, igniting a chain of events that would lead to the term “blockchain” becoming common parlance in discussions about technology in the corporate world. Over the subsequent decade blockchain, the technology underpinning Bitcoin, inspired a multitude of other cryptocurrencies, and numerous technology solutions. As the price of Bitcoin itself was seemingly headed “to the moon” — peaking at a vertiginous $19,783.06 USD — blockchain was being touted as the solution to an increasingly diverse range of problems. A combination of two factors — the massive hype generated by the lofty valuations of cryptocurrencies, and the relatively esoteric nature of the technology and mathematics involved in blockchain — led to the label “blockchain” being linked to technologies that were arguably (or in many cases, blatantly) not blockchain.
Ash Priest, managing partner at Novigi, recently chaired an innovation discussion group on behalf of the Association of Superannuation Funds Australia (ASFA) to discuss blockchain and its applications in superannuation. Around 25 thought-leaders and decision-makers from in and around the superannuation industry gathered to flesh out exactly what the applications of blockchain might be in their field.
What is blockchain?
In a sentence:
There are countless detailed descriptions of the workings of blockchain available on the internet. We actually believe that the original whitepaper is still one of the best descriptions of how blockchain works in practice. Rather than rehashing this (get it?), we focus here on a few key concepts and definitions that we believe must be grasped to be able to think critically about how blockchain and related technologies might be applied in a superannuation context.
Distributed Ledger Technology (DLT)
Blockchain is a specific example of a distributed ledger technology (DLT). In other words, all blockchain is DLT, but not all DLT is blockchain. The defining characteristics of a DLT are:
Each “node” (computer participating in the network) in a DLT holds a copy of the entire ledger, that is, each participant in the network will have access to every record created since the network’s inception
Some form of cryptographic digital signature will be used to verify that a record is indeed associated with the parties it purports to be
Some form of consensus algorithm will be used to ensure that all participants in the network accept the same set of facts as truth.
To be fairly characterised as a blockchain, a DLT must contain some concept of blocks — groups of records/transactions containing a cryptographic hash of the previous block, along with a timestamp — the veracity of which are periodically agreed upon by the network as a whole.
Public blockchain and other DLT solutions are “pseudonymous”. This means that, contrary to popular belief, blockchain technologies such as Bitcoin are not truly anonymous. Rather, anyone viewing the Bitcoin blockchain would be able to identify all transactions associated with a particular user’s pseudonym — in this case their public key. The degree to which Bitcoin is anonymous is thus a function of a person’s ability to keep the association between their public key and their real-world identity a secret. Considering that anyone you transact with will have access to your public key (think: your employer, your local coffee place, any friends you transfer money to), this ability is probably extremely limited.
What blockchain and DLT do particularly well is make it difficult for malicious actors to assume your identity. They do this by using public-key cryptography to implement digital signatures. Each user of a DLT will have both a public and a private key. With these the user will be able to leave a digital signature (usually in the form of a hexadecimal hash) on a record/transaction. From this, any other user in the network can verify — by looking solely at the signatory’s public key, signature and the content of the record/transaction — that they do in fact possess the private key. This is despite the fact that they never have knowledge of the signatory’s private key. This is made possible through some rather involved mathematics (the details of which can be found here).
As blockchain and DLT solutions are, by nature, distributed, there needs to be some method by which disparate nodes reach a consensus on what set of records to accept as the truth. Having some form of consensus algorithm also ensures that new records can be reconciled with existing records in the ledger. This safeguards against double spending and other logically similar issues that can arise in transactional systems.
Bitcoin, for example, relies on a “proof-of-work” algorithm for establishing consensus. In this model, nodes are incentivised to verify the validity and chronological provenance of transactions. This is done by awarding nodes with Bitcoin for solving computationally intensive mathematical problems, i.e. creating proof-of-work. The network then accepts the version of the truth that the most computational power has been expended on verifying, that is, the blockchain with the most extensive proof-of-work. For malicious actors to enforce an alternative version of the truth, they must have control over a majority of nodes in the network. Provided that this is impractical — i.e. the number of nodes are sufficiently high and the nodes themselves independent — the chances of an attack like this succeeding are negligible. This is a key benefit of blockchain as a technology.
What does this mean for Superannuation?
The Australian superannuation industry has been discussing the potential applications of blockchain for the better part of the last decade, and has been subject to the same hype and confusion that has characterised blockchain more broadly. Recent developments in the use of blockchain and other DLT in enterprise contexts — coupled with what we hope was a level-headed appraisal of the technology’s capabilities — allowed us to cut through to the heart of the issue in the recent ASFA innovation discussion group. The key area that we identified blockchain and DLT as having a viable use-case in superannuation was identity, and it was also briefly discussed as a potential replacement for traditional registry systems. There was broad consensus in the room that identity would be a sensible “first cab off the rank”, and so we go into some detail on this topic here.
Superannuation is a popular target for fraudsters and con artists. The Australian Criminal Intelligence Commission (ACIC) identify superannuation fraud as one of the main categories of fraud perpetrated in Australia, stating that “fraudsters…steal superannuation-related correspondence from mailboxes, or access personal information on the internet, to assume the superannuation fund member’s identity and gain access to their funds”. Given the large sums of money involved in superannuation ($2.9 trillion at the end of the June 2019 quarter), fraud of this kind has a large impact on individuals, and indeed on the economy as a whole.
An identity solution built on digital signatures and using a blockchain solution as a distributed record of identity information could make it significantly more difficult for malicious actors to commit identity fraud in superannuation. A number of solutions of this kind exist at various stages of development — notable examples include offerings from Microsoft and IBM respectively. In essence, the high-level mechanics of these identity solutions are the same. A user has a public and private key, and can use these to digitally sign transactions, smart contracts and other kinds of records. A blockchain stores the association between a user and their public key, which is verified using a consensus algorithm.
We see such a solution being adopted by the superannuation industry in one of two ways. Firstly, as a government mandated initiative, likely implemented and governed by the ATO. One could easily imagine a future in which a member was issued with a public and private key by the ATO (perhaps on filing for a tax file number), which was linked to a person’s real-word identity by a record stored in a blockchain accessible to all funds. Alternatively, a member’s first super fund could be responsible for issuing that member with a key pair that they retain for the rest of their life. In both scenarios processes like rollovers and withdrawals would involve the member digitally signing the request, and any fund parties involved verifying the digital signature against records held in a blockchain. This process would both substantially reduce the risk of identity fraud, as well as reduce the overhead associated with identity verification.
There are, of course, some potential barriers to adoption here. Failing compulsion by government, the reluctance to be prime-movers among funds is a key issue. This is compounded with solutions like blockchain, which depend heavily on network effects to provide any benefit, and indeed to function at all — a consensus algorithm means very little in a network with a single node.
There are also issues that need to be solved around consensus itself. In a cryptocurrency, nodes are incentivised to verify transactions and come to a consensus by offering financial rewards in exchange for computation. What this would look like in a network where super funds are nodes verifying identity information needs some thought.
Finally, key storage presents some logistical difficulties. A member’s private key would potentially become an extremely important secret to be protected. Many solutions for key storage exist currently, from local software storage, to hardware solutions, to offerings from major cloud platforms (AWS, Azure, GCP etc.). To minimise risk here, we believe it would be prudent to take a uniform approach to this problem, rather than to leave it to individual members — lest fund call centres be inundated by members who have misplaced their private keys.
Earlier this year The Economist wrote that “there is so much hype over organisations using blockchain to ‘solve problems’ where ordinary databases or non-blockchain systems would do, that it is hard to appreciate its real importance.” We very much agree with this statement, but continue to believe that blockchain can play an important role in the technological ecosystem of funds in years to come. On the proviso that either funds coordinate and collaborate to innovate in this space, or that government and regulation show real leadership, we are optimistic that we will see blockchain in superannuation in the not too distant future.
Kevin Fernandez leads the consulting business at Novigi, and is based in the Melbourne office.
For more information about anything you’ve read here, or if you have a more general inquiry, please contact us.