What happens when an unstoppable force meets an immovable object? The unstoppable force is customer expectation of real time, which we get from using email, text and social media. We are instant gratification junkies, like Veruca in Willy Wonka “I want it and I want it now”. The immovable force is the batch legacy that derived from IT 1.0 moving from paper systems to computers about half a century ago.
Expect a loud noise!
Personally I am betting on the unstoppable force. You don’t go far wrong by serving customer needs. If Veruca wants it now, find a way to sell it to her!
First we look at the 5 batch hangovers from paper. Then we look at the responses of different players in the ecosystem.
The 5 batch hangovers from paper
- LIBOR & other “Fixes”.
- Quarterly Financial Reporting
- Daily Accounts Close
- T+2/3 Securities Settlement
- Regulatory Reporting
LIBOR & other “Fixes”.
Quick word association – LIBOR = scandal (if you want to understand the LIBOR FX scandal, go to this BBC explainer). Another batch “fix” (when a price is fixed for the day in a process dating back pre computers) is for Gold (and other precious metals). The gold fix used to happen in the physical offices of Rothschild and then, in a classic bandaid move, became updated by moving to a teleconference system (this Wikipedia article explains the history). The point is simply that when 21st century real time global trading in FX and precious metals meets a Victorian batch hangover from paper, there is a lot of room for collusion and other illegal activity. The regulatory response so far has been to layer more processes on the banks so that their compliance people can catch the bad traders while they are colluding via phone, email or trading terminal messaging system. Lets apply more bandaid! Bandaid vendors vote for this response.
A more radical response would be to replace the fix with a computed “daily price”. This is technically simple. There are multiple methods (eg. trade weighted average) and banks that need a daily closing price (eg to value a portfolio or an index) can choose which algo they use (all of them would be open source). If one algo gets gamed too much, they can switch to a different one (which will reduce the value of gaming the system). Our trading systems will cope with this quite easily. The problem is which regulator in which country can make this change? This is difficult because it will affect banks and traders in every country.
Quarterly Financial Reporting
Our systems record transactions as they happen. It would be technically simply to apply a filtering process in order to issue an abstract of that data (not so much that it gave away trade secrets) to investors in real time. Instead we have layers of processes to produce a report for investors and regulators on a quarterly basis. Plus we have a whole industry of analysts and media whose job it is to guess the numbers and to manage investors expectation of those numbers.
Core Banking Daily Accounts Close
Little known fact – CORE stands for “centralized online real-time exchange”. What we think of as those old-fashioned legacy systems were disruptive in the 1970s when we moved from paper in bank branches to branches that were connected via a network to a head office where all the accounts were processed by a computer. This allowed branches to exchange account data in real time. The paper systems had been processed quarterly (try doing it any quicker with paper). Core banking systems speeded that up massively by going to daily close. It is still batch – just much faster batch.
In an era when we debate the value of branches vs mobile phone banking, this automation of accounting between branches seems rather quaint.
T+2/3 Securities Settlement
This is where the Ferrari meets the Donkey. We measure trading in fractions of a second. That is the Ferrari. Then we move to Settlement which takes 2 or 3 days (T+2/3 Securities Settlement). This is starting to change, enabled by Blockchain technology (for an analysis of three projects in this area read this post).
Regulators demand data from banks and traders. In order to not be too intrusive, they only ask for that data in predefined periods – days, weeks, months, quarters, years depending on the data. Banks have to reconcile what they send with what goes to investors and internal management. Any mistake and the fines come pouring in. Pity the poor compliance officer.
What if all the data was streamed in real time (after a suitable filter to protect customer confidentiality and trade secrets)? Regulators, investors and internal management could all parse the feed to get what they need for their decision support systems.
Responses from different players in the ecosystem
The reason change takes so long is that no single entity can make a decision in isolation. There are too many moving parts and as any deal-maker knows, when there are too many moving parts, nothing happens. Each of these players in the ecosystem have a different response to the unstoppable force of real time.
- Incumbent Financial Institutions
- StartUp Neobanks
- IT Vendors
(Regulators also need to respond but mainly to get rid of the vestiges of Victorian days via daily fixes as described above).
The Incumbent Financial Institution Response
When banks study their legacy systems, which they are maintaining at huge cost, they discover massive overlap with many systems serving exactly the same need. The sensible reaction, to get rid of the overlap, gets crushed by a business group telling IT that their business is utterly reliant on that legacy system and suggesting that any glitch in the conversion would be career suicide for the project manager. In one case, a system had only one user!
This has to change. The question is how. There are two corporate responses:
- Lets do a total core system revamp, move everything to real time. It will take years and cost millions but the effect on our agility and sustainable competitive advantage will be a game-changer.
- Lets bypass the old legacy systems (let IT move them to “end of life maintenance” when the time comes) and simply buy the startups when they get to reasonable scale or partner with startups or other Banks by using Open APIs on top of utility platforms.
The Response of StartUp Banks
StartUp Banks (aka as “challenger banks” or “full stack startup banks” or “neobanks”) are fully regulated Financial Institutions that don’t have any legacy. Its that simple. They can meet Veruca’s demand for real time, because everything they do is in real time. They still need some batch jobs that are designed to serve the needs of investors and regulators, but those batch jobs are simple after the fact reporting; the cart does not drive the horse.
As these StartUp Bank start to get actual market share, it will impact the Incumbent Financial Institution Response (decisions where they assumed they had plenty of time now become a lot more urgent).
Traditional Fintech is a huge business (see this report, its a few years old but this is not an area of rapid change). It may not be “exciting” but today it is bigger than Emergent Fintech and, done right, it is hugely profitable.
There are three responses:
- the Banks are locked into our legacy systems and its a massive cash cow, lets milk it for everything it is worth. Private Equity loves this approach as it can service big debt loads.
- lets be the vendor of choice when the Banks need a transformational move to real time. This is only possible for a few very big vendors. There are not that many deals like this but they are huge.
- forget about the legacy systems, we are about new systems that “sit on top”. That has been the de facto response for decades. The problem now is that we are reaching the limits of what we can do on top. If we offer a real time response to the customer but the transactional processing is batch, there is room for a lot of error and liability.
Most IT Vendors have a mix of these three responses, but the emphasis and mix varies a lot.
Expect this subject to be debated vigorously in a few weeks time in Geneva when 8,000 bankers and their vendors arrive for the annual gathering of the tribes known as SIBOS. See y’all there.
Daily Fintech Advisers provides strategic consulting to organizations with business and investment interests in Fintech & operates the Fintech Genome P2P Knowledge Network. Bernard Lunn is a Fintech thought-leader.
Like This Post
Original Post: https://dailyfintech.com/2016/08/05/real-time-banking/