Innovation

In the ledger we trust

South African banks are dabbling with blockchain and distributed ledgers, learning some important lessons along the way.


1 March 2017
Brought to you by
(Left to right) Waseem Hassim, Nedbank, Tanya Knowles, Fractal Solutions, Farzam Ehsani, RMB (Karolina Komendera)

The adoption of blockchain is a fraught one, perhaps because we’ve been down a similar road before.

Back in the 1970s, before the internet as we know it today, engineers started work on a protocol for the architecture of networking systems.

They came up with something they called the Open Systems Interconnection (OSI) Model, intended to govern the protocols used by international internetworks. The development of OSI was slow, however, and there was also another simplified standard of OSI called TCP/IP, or Transmission Control Protocol/Internet Protocol, developed by engineers Bob Kahn and Vint Cerf. By the 1990s, ethernet and TCIP/IP had become the networking standard.

For Peter Munnings, a developer working blockchain at Rand Merchant Bank, this tussle between protocols looks all too familiar.

“If you’d asked those engineers what they thought the future of this massive network they were dreaming about would look like, none of them would have said the killer application was going to be people sharing pictures of their children and dogs. They’d have probably said it was going to be a financial application.”

Blockchain technology is powerful because it enables transactions to take place without one party having to trust the other. It also means there’s no need for another organisation, such as a bank, to be involved. It’s mostly tamper-proof and once a transaction has been entered on the distributed ledger, it’s there for good.

Munnings, along with another developer, Coenie Beyers, and team leader Farzam Ehsani, are the only three people employed full-time in the South African banking sector to work on blockchain. They work at the Foundery, an apt enough name for a place where new ideas are forged, and sit at three desks in a plush open-plan office in a three-storey building on the RMB campus in Sandton. Blockchain skills are exceedingly rare, and Beyers and Munnings have since left the bank and are now working at decentralised app studio ConsenSYS.

Future of banking

Munnings says the Foundery mines ideas that emerge from within the bank, ‘to see if they have any legs’. “And then we try to build a prototype. We’re trying to figure out what the future of banking looks like and how best we can prepare ourselves for that.”

Ehsani describes his job as being the interface ‘between the technical and the non-technical’.

For him, no one knows what the future of blockchain will bring, ‘but we know the result’.

“With the internet, we knew that communication was going to become more efficient and cheaper. I can tell you that banking in the future is also going to become more efficient and cheaper. What the exact token will be, the apps, who the players and the institutions will be, none of us can tell you that.”

Waseem Hassim, Nedbank’s group technology strategy execution executive, says the board and CIO are committed to considering all application of blockchain technology. "The objective is to leverage what is currently seen as a disruptive technology for business benefit and value delivery," says Hassim, adding they were at an 'advanced stage' of completing technical impact assessments across a number of internal use cases, and plan to have a 'tangible deliverable' by the middle of this year.

"We are starting out with a small use case and intend to grow organically as we decipher our way through the hype." he says.

That hype has now reached fever pitch. The world over, bankers, lawyers, insurance brokers and developers, among others, are toying with blockchain technology. For organisations that manage other people’s assets, the promise ‒ and some would say threat ‒ of blockchain is that assets can be managed in a decentralised way, profoundly affecting long held business models.

Profound, powerful

For Andrew Baker, CIO for corporate and investment banking at the Barclays Africa Group, the hype was there for good reason, as people were seeking to attract venture capital.

Baker says he’s ‘more of a technical person’ and before his current job, he spent a decade developing high-frequency and algorithmic trading systems.

While he doesn’t mention it when speaking to Brainstorm, other bankers say he was among the first to start discussing blockchain across the industry. One banker calls him a ‘visionary’.

Baker says he isn’t seeing a lot of maturity in any particular blockchain stack at the moment, but some of the ecosystems have quite large communities, echoing the sentiment that mass adoption usually wins, despite the elegance of the code.

He also doesn’t see blockchain having a particularly long life, despite it being a ‘stunning’ technology.

Distributed ledgers, however, are ‘really interesting, profound and powerful’, and some of the emerging technologies feel to him to have a measure of maturity.

At the time of writing, Barclays Africa is the only South African bank to sign up with the international R3 consortium investigating blockchain use cases. The consortium is working on a distributed ledger platform called Corda, specifically designed for financial services. Baker says a couple of his developers are in London working on the Corda syndicated loan platform.

He adds that technologies developed outside of banks often do something different ‘to the way banks work’, such as not having privacy restrictions, partitioning or regulatory nodes. Corda’s architectural principals are markedly different, however, and much more suited to banking protocol, so much so that “we could use that technology between me and several other banks to perform bank-to-bank transactions. And I see that happening sooner rather than later,” he says.

Sketching how he sees distributed ledgers and blockchain affecting financial services, Baker says there are two worlds: bank-to-bank and customer-to-bank. In the case of the former, he says that right now, they are operationally ‘not totally awesome’.

Stock loans, for instance, are still done over the phone, and this is a use case that can benefit from automation and its resultant efficiencies.

In the bank-to-customer arena, there’s huge potential for disruption, and there are many players that could get ‘bumped out of the ecosystem’, like card providers or any organisation that performs routing between a customer and a financial service. In fact, anything that happens between a financial institution and a customer, such as Know Your Customer (KYC) or payments, is ripe for disruption.

By comparison, a bank-to-bank agreement is straightforward and can be settled over coffee with another banker. A widespread bank-to-customer scenario is more like a community decision, and would, of course, invite regulation.

Baker sees less regulation for banks using distributed ledgers to send one another messages. In this case, the asset is not actually being stored on a ledger, but remains with the banks.

If all the banks agree to expose identical APIs to one another, real-time clearing or real-time gross settlement becomes possible.

A delicate egg-dance

The net effect of all this means that today’s banking model is on notice, and bank fees are set to decrease, or fall away completely.

As one banker puts it: “It’s like a turkey voting for Christmas.”

Now this is just the kind of dangerous talk that can get a banker into serious trouble, “but more and more central banks are taking (blockchain) very seriously,” says RMB’s Ehsani, running off a string of names including the Bank of England, the Bank of China, the Bank of Canada, the Monetary Authority of Singapore and Hong Kong Monetary Authority.

“You name it. Even our central bank has put on record that it is looking into blockchain, cryptocurrencies and a regulated crypto rand.”

The key words here are ‘looking into’.

The bankers and regulators are involved in a delicate egg-dance. Everyone speaks in terms of a ‘journey’ and ‘learnings’ and ‘sandboxing’.


We’re trying to figure out what the future of banking looks like and how best we can prepare ourselves for that.

Peter Munnings, Rand Merchant Bank


Meanwhile, at present, Baker says there are just too many actors in the way of a simple transaction.

“If I go to a point of sale and I have a current account, at the moment there’s a four-box model (the payer and their bank, and the seller and their bank).

“There are all sorts of people taking money for different services. There’s also a lot of fraud and the customer experience is not great. If we landed on another planet today, and we wanted to design a payment system, it wouldn’t look anything like it does today. We could do a fraud-free, point-to-point transaction for payment that would utilise something like a distributed ledger.”

One of the side effects of financial institutions investigating blockchain is the way they’ve had to change their mindsets around collaboration.

Sharing is caring

While some bankers had started talking among themselves, Strate, South Africa’s central securities depository, early last year offered to chair the discussions.

This was easier said than done, says Tanya Knowles, the managing executive of Fractal Solutions, a division of Strate, and often it became hard to focus discussion topics because of the size of the group.

“Some people are interested in payments, some in trade finance, securities, or syndicated loans. Our attitude is what you put in, is what you’re going to get out of it. So if you’re passionate about a particular use case, then you drive that forward.

“It’s driven on a case-by-case basis. You can come and present your idea to the group and see who’s interested, and then off you go.”

In an attempt to bring some structure to the group, it was divided into three streams: regulatory, technical, and education.

They’ve also tried to structure the group using blockchain’s collaborative principles, so there’s no single person or group in charge.

“If we can brainstorm what a strategy for the overall market will look like then we can start to make decisions. We don’t want to be dictated to from the top down, without us being involved,” says Knowles.

For Baker, ‘the most profound thing that banks will see is running a shared infrastructure’.

It’s a sharing culture, and that’s tough for us banks because we never did that before. We have to share code, we have to share ideas, and we create shared products, and I think a lot of banks really struggle with that conceptually. We are born to compete with each other.

“And the beauty of doing it is that we get to learn.”