June 30, 2016
The blockchain technology, which is the distributed ledger foundation behind the ‘bitcoin’ cryptocurrency, has already been recognized as a potential enabler for the accelerated velocity of money in almost every sector of the financial industry. It is certainly an interesting technology platform worthy of studying and understanding, exploring and experimenting with due to its underlying immutable storage and ability to transparently distribute transaction data structures between network participants.
As such, the blockchain provides a potential path for legacy financial systems to interoperate, greatly improving the efficiency of their current processes. Central banks, financial institutions and technology firms are all taking a serious look at the various blockchain platforms, evaluating potential use cases and also learning about its potential and challenges through the ongoing PoC projects.
In the payments space, for example, blockchain could enable efficiency improvements of the clearing and settlement in the 4-party payment ecosystem (enabled mainly by Visa, MasterCard-based networks). I have to be very clear right from the start – the solution I am going to present and propose here doesn’t suggest in any way eliminating the role of the central payment networks in facilitating the execution of the payment transactions.
My goal is only to suggest the different ways for how the current 4-party payment network’s clearing and settlement process between issuers and acquirers could be facilitated more efficiently through a common blockchain layer with the payment network still playing an important (but somewhat modified) role. That ultimately should result in the elimination of the need for costly end of day file reconciliations, delays due to detected inconsistencies and allow for the acquirers to consistently pay merchants on the very same day when they made the sale. Besides giving acquirers opportunity to pay merchants on the same day, the improved efficiency of the payment clearing and settlement processes may also reduce the payment networks’ and card issuers’ operating costs and neutralize further negative impacts of slashed interchange fees on their bottom lines.
Let’s dive right into the use case and take a look at how Visa and/or MasterCard (applies to Interac and other national debit schemes as well) may potentially utilize blockchain technology in order to enable near real-time clearing and settlement of funds between card issuers and merchant acquirers. I will try to stay at a high level and not suggest any particular blockchain implementation platform. I believe that the concept is fairly general in nature and as such, should be implementable on any of the popular basic blockchains.
I will just assume, for pure reasons of scalability that the underlying blockchain platform should be ‘permissioned’ with participation limited only to payment networks and acquirers and card issuers, which are payment network association members. It is also reasonable to assume that in such a blockchain network, the consensus can simply and reliably be achieved just between the single merchant acquirer and the single card issuer, which were involved in each payment transaction execution. Clearly, there is no need to involve all the acquirers and the issuers in vetting the payment transaction that involves a single acquirer and a single issuer since that would unnecessarily complicate and slow down the underlying consensus reaching process without adding much value. No proof of work, no proof of stake, no need for complex consensus algorithms. The set of payment network ‘mining’ nodes would simply listen to the already vetted blockchain transactions between issuers and acquirers and group them into the final blocks, supported with very efficient underlying consensus. With these design assumptions, the overall performance of clearing and settlement blockchain can be kept equal (or fairly close to) the performance of the existing payment card authorization transaction executions. Things should be kept as simple as possible to gain performance but without sacrificing immutability and security.
Let’s also assume that the payment network is fully in charge of issuing its own cryptocurrency, which just represents the underlying asset (token) pegged to the fiat currency for facilitating intra-day, near real time clearing between issuers and acquirers through the blockchain. Let’s call such cryptocurrency ‘pnCOIN.’ In the case of Visa, it may be called ‘vCOIN’; in the case of MasterCard, it may be called ‘mCOIN’; Interac may call it ‘iCOIN’, etc. No blockchain miners (like in bitcoin or Ethereum implementations) would be necessary in such an implementation.
At the beginning of each business day (whatever that is in any particular instance) every participating card issuer and merchant acquirer would ‘purchase’ a certain amount of pnCOIN cryptocurrency from the payment network. Basically, the payment network submits the set of blockchain transactions, which transfers the ownership of the purchased amount of pnCOINs from itself to each acquirer and card issuer. This way, each acquirer and issuer prepares itself for the near real-time bilateral clearing of the daily payment transactions through the underlying blockchain. Let’s call this process of converting the amount fiat currency to the pnCOIN amount ‘Cash-OUT’. The received money from participating issuers and acquirers is held in the special operating account under the payment network control.
When the payment card (or NFC) transactions happen during the day (in store or online), they would normally be authorized through the regular payment network infrastructure, as today. Nothing changes in that process. However, the card issuer and acquirer that were involved in the original payment transaction now agree on the separate ‘purchase clearing’ blockchain transaction, which is submitted to the ‘clearing’ blockchain network by the card issuer. The ‘purchase clearing’ blockchain transaction simply transfers the ownership of (transaction amount – applicable fees) of pnCOINs from the card issuer to the acquirer. As a result, the acquirer’s individual UTXO is increased and the issuer’s individual UTXO is decreased by the same amount of pnCOINs. This process of transferring (transaction amount – applicable fees) of pnCOINs from card issuers to acquirers is repeated for every payment transaction between any issuer and acquirer during the business day.
The ‘chargeback clearing’ blockchain transactions (as initiated by issuers) would result in the reverse transfers of a certain amount of pnCOIN ownership from the acquirer to the issuer and would be represented with their own ‘clearing’ blockchain transactions. Similarly, ‘reversal clearing’ blockchain transaction would exist for clearing of reversals and would be submitted by the acquirers, etc. All these ‘clearing’ blockchain transactions are happening in parallel with (and as result of) the normal payment authorization/chargeback/reversal transactions during the day.
At the end of the business day, each acquirer and card issuer can now ‘Cash-IN’ their accumulated pnCOIN totals (i.e. individual UTXOs) with the payment network. The ‘Cash-IN’ process effectively transfers ownership of the outstanding pnCOIN UTXOs from each of the issuers and acquirers back to the payment network. In return, each acquirer and issuer would receive the equivalent amount of fiat currency from the payment network. In an ideal case, at the end of the ‘Cash-IN’ process, each individual acquirer's and issuer's pnCOIN UTXO 'balance' would drop to zero, and also, the payment network’s operating account balance should become zero. Based on the received funds from the payment network, each acquirer can immediately pay their merchants for the sales they made (of course minus applicable merchant discount rates).
Because of the immutable nature of the underlying ‘clearing’ blockchain distributed ledger platform (with very efficient bilateral consensus between transacting acquirers and issuers), the need for end of day clearing and settlement reconciliation between acquirers, issuers and payment network is clearly eliminated. The end of day state of the clearing blockchain distributed ledger should be considered final and already agreed between the transacting and participating parties.
This implementation assumes only basic blockchain platform capability (basic ‘bitcon like’ asset transfer capability backed with much simpler and efficient consensus algorithm), without any requirements for the complex, immature and still error-prone Turing complete and general-purpose ‘smart contract’ programming language environments.
This is just one example of how basic blockchain capability could potentially be practically used by the payments industry to improve the efficiency of the current transaction clearing and settlement processes to eliminate need for any type of costly end of day reconciliation between participants. Of course, this description isn’t a complete solution by any means since it doesn’t obviously take into account clearing and settling of multi-currency, cross border transactions, but nevertheless, it can be used as a basis and expanded to cover more complex real-life situations.
This same model can also easily be adapted for and applied in a similar fashion for the post-trade clearing and settlement in capital markets (OTC trades come to mind immediately here) with CCP replacing the role of the payment network.