Tuesday, 24 March 2015

BBY Disruptive Lunch: CBA's Kelly Bayer Rosmarin

It's sometimes easier to write about the "less" polished presentations we get at the BBY Disruptive Lunches then those that are focused and tested. Commonwealth Bank's (CBA) Kelly Bayer Rosmarin presented a cogent and polished summary of that institution's approach to disruptive technology generally and specifically to the digital client interface.

Bayer Rosmarin's engineering background was to the fore as she set limits on the subjects she was going to discuss and stuck to those parameters. Part of the charm of these lunches is the way that some of the start-ups meander through their business cases almost rethinking what they're saying as they go. I don't expect young companies to have laser sharp focus. When they do, as was the case with someone like Catapult Sports it can be riveting, but it's not always necessary. Take Pocket Book's (19  Feb 15) offering last month, that to me was notable because I got the sense the team was still discovering their capabilities and a best avenue of action.

Bayer Rosmarin sought to deal with disruptive technology and CBA through the lens of three distinct examples: Crypto Currencies, Cognitive Computing and Cyber Crime.

1. Crypto Currencies

Would crypto currencies exist without the 2008 financial crisis? Security, trust, convenience and efficiency, all areas severely tested by the crisis. The collapse of major banks severely tested the faith of people during the GFC. Historic examples from past economic adventures (think Weimar Germany, South American banana republics, African hyperinflation) provided fertile ground for Bitcoin (and other crypto currencies) in recent years.

Bayer Rosmarin's Bitcoin case study was a gentle debunking of the currency.

Firstly, coming from the angle of a large bank the security aspect of Bitcoin seemed to CBA somewhat counter-intuitive. Approximately 10% of Bitcoins would never be used because the security system functioned on the basis of a one-time passcode. Lose that, and you lose your ability to redeem your "money". Imagine (she suggested) if CBA froze your account permanently if you lose a passcode. The bank would likely to suffer an outflow of customers. How then could Bitcoin pass a convenience test?

Secondly there are a limited number of Bitcoins. I was somewhat confused by the CBA argument here because as we know Bitcoins can be split into fractions down to (I believe) nine decimal places. Additionally the raison d'etre for most Bitcoin believers is that the cap on the eventual size of the Bitcoin pool serves as a hedge against their concerns about central bank incompetency and associated currency debasement.

Thirdly, and more convincingly Bayer Rosmarin pointed to concerns as to the security of Bitcoin exchanges given various scandals that have erupted in the last two years. It's hard even for ardent Bitcoin enthusiasts to rebut this because the evidence is strong that this has been the weakest facet of the currency.

Lastly, Bitcoin according to CBA lacks differentiation. In fact, anyone can start a crypto currency. I think the suggest here was that these currencies might sink or swim as a group. If one currency was exposed as having certain vulnerabilities, then the group was likely to suffer because the flight to safety element of say out of Euro and into Swiss Francs wasn't available.

Not everything about crypto currencies is negative. Various protocols for payments were worth exploring. CBA have recently paid $A40 million for South African-based Take Your Money Everywhere. "TYME" is one of those bridging technologies that gives lower end or under-serviced users access to regulated bank accounts. The protocols and tech derived from this business is likely to add growth for CBA across not only Africa, but also in Asia. Bayer Rosmarin's analysis is clearly concluding that crypto-currencies are likely to stay as niche speculative ventures if banks can link their services in a way to make their day to day use unnecessary.

2. Cognitive Computing

The roots of today's Cognitive Computing can be traced back to the work done by IBM's lab in the mid 90's. That research culminated in the famous series of chess matches between supercomputer Deep Blue and Garry Kasparov.

Fast forward to today and think self-driving cars and algorithmic securities trading. The key to each step in these developments was the ability to handle ever-increasing amounts of data. "Big Data" as readers will know is an ever present theme at these lunches. For institutions such as a big bank, the question becomes one of sourcing and distributing access to that data, and what to charge or pay.

If you're a CBA SME customer, you can access some of that data through their Daily IQ application.

As a further add-on, CBA offers an open interface called Pi. The front-end of Pi is "Albert" and that gives you the opportunity to develop your own ways of interacting with clients.

Interestingly I allowed my mind to stray to some of the many presentations we saw last year which purported to provide just this type of add-on to users. As a friend of mine who runs a restaurant said when I showed him Posse (https://posse.com): "Why should I pay for that? I think my bank already offers the same thing?" What it says to me as an investor is that many of the businesses I look at are predicated on a single business plan. Attract a buyer such as CBA before they put their mind to developing their offering in your chosen sector. I know that sounds obvious. I've lost count of the number of times I've walked out of presentations thinking I've just seen the next unicorn only to change my mind after a cooling off period.

3. Cyber Crime

Readers will know I'm a long time fan of predictive algorithmics. We saw these methods first adopted by the Israelis in order to be able to deploy their limited security forces to the maximum effect. Financial institutions are already using these same techniques to track fraud and security infractions within their businesses. Cybercrime remains a huge challenge. Biometrics is often put forward as a solution, but Bayer Rosmarin doesn't see this as a standalone answer to tech-crime.

If you can find a business with a unique slant on cyber crime, I'd grip on tight. The very fact that there is no current industry standard solution suggests to me that this is an area that investors have a chance to make serious returns. Maybe part "Local Measure", part biometrics is what I'm looking for to fill the gap?


I view Kelly Bayer Rosmarin's presentation as a nice counterpoint to much of what we've been privileged to see and hear over the last year of BBY Disruptive Lunches. Call it the "Empire Strikes Back" if you will. It was at its core a well reasoned and somewhat sobering reminder that big institutions are well able to fill niches that smaller more nimble entrepreneurs are claiming as their own. One question during the Q&A best summed this up. Bayer Rosmarin was asked about the challenge being made to CBA's business by peer to peer lenders. While not dismissing the sector out of hand, she did suggest that operators (e.g. Society One) might find their businesses struggling if interest rates started rising. If you think about it, that makes sense. P2P's model is based on occupying the spread between lenders and borrowers. As rates rise that spread is likely to narrow and with it squeeze P2P business. Ultimately I got the sense that Bayer Rosmarin modus operandi is not to jump at shadows. Like all good engineers, it would seem she prefers to analyze first and act after carefully weighing up the possible responses and solutions.


No comments:

Post a Comment