The disclosure by Nedbank of metrics in its retail banking division is, quite simply, industry leading. I was struck by this – again – while reading its interim results presentation and booklet yesterday.
One might argue that it has to share this amount of detail, given the multi-year “evolution” underway in the unit. The latest is part of its “target operating model” programme. By 2020, it aims to lower its efficiency ratio (cost-to-income) for retail and business banking (RBB) at equal to or below 58%, from 64% currently. Because of its relative size (and large fixed cost base), Nedbank has historically struggled with a stubbornly high efficiency ratio. Contrast this with Absa Group’s local RBB operations at 58.3 (from 56.6 a year ago), and one can see the disparity.
Nedbank needs to get the retail and business banking unit’s efficiency ratio to under 60 in order to achieve a groupwide target of ≤53% by 2020. It continues to achieve cost savings, and banked R237 million worth in the unit during the first half of 2018 (R177 million related to its target operating model efforts). Space reduction in the branch network continues apace – it discloses the cumulative square meterage saved – and it has reduced headcount by over 1 000 staff in the past 18 months, mostly through attrition. It even provides a breakdown of ‘new image’ versus ‘traditional’ branches as it continues the network modernisation.
But it’s in the client data where Nedbank really stands out from the rest. Apart from disclosing total retail clients, transactional ones and main-banked customers (which it always has), the bank has introduced additional metrics: active clients and consistently main-banked clients.
The former is any client who has been active in the past six months, while the latter is a main-banked customer for each of the past 12 months. Additionally, it has provided further details on what constitutes a main-banked customer. In other words, this would be how many transactions customers in different segments need to perform over a three-month period. This is an unprecedented level of disclosure.
We also know how many customers it has in each of its segments (middle vs professional vs entry-level and so on), as well as how many retail customers it has in each product line (transactional vs home loans vs card, etc.). For its lending product lines, it publishes the number of applications received as well as product-specific metrics, such as the average loan size in personal loans and loan-to-value of new business in home loans.
Capitec has also improved disclosure significantly, now distinguishing between credit clients, transact and saving clients, as well as primary bank customers. In April, Capitec introduced another metric: “quality” banking clients, which it defines as “those clients who have stable inflows into their account and stable product usage over a consecutive three-month period” (read: Capitec execs surely still smiling about this number…). In its lending business, Capitec provides superb disclosure, especially when compared to other lenders (Viceroy may disagree).
Standard Bank also provides a fairly detailed breakdown of customers, with splits for credit card accounts, current accounts, targeted current accounts and so on. It provides some level of disclosure on trends within its lending products, such as home loans. Standard metrics for its distribution network as well as digital customer behaviour are disclosed.
FNB parent FirstRand publishes customer account figures once a year, in its integrated report, with selective trend disclosure for segments at the interim and full-year stage. Also, the bank discloses transaction volumes across its various channels, including digital.
Contrast this with Absa Group, which discloses so little about its retail banking business that it’s laughable. From its interim results booklet, we know that the “average number of products per customer increased marginally” thanks to a product penetration chart. We also know the annual loan production growth per segment (card/home loans/vehicle finance), which we’d have been able to ascertain from the segmental breakdown in its financials.
And we know how many Absa Rewards customers it has (2.8 million, versus 2.6 million a year ago). Like all other banks, it publishes its distribution footprint (but, with no detail). That’s it.
Since last June, it has not published retail banking customers (and even then, it made very selective disclosure).
For years, it’s been trying to fix its retail and business banking unit in South Africa. Craig Bond, hired from Standard Bank about five years ago, couldn’t and was shifted to a token role (CEO of partnerships, joint ventures and strategic alliances), before stepping down last month.
The latest strategy is focused on “regaining market leadership by leveraging the scale of the business through focusing on consumer finance whilst deepening the relationship with customers to benefit the transactional franchise”. As I pointed out in April, this is no real change from an ambition articulated by Ramos in 2014, when she stated that the bank wanted to regain the leading market position in retail and business banking.
Four years on and not much has changed.
Surely investors ought to demand better disclosure for a unit – transactional and deposits – that delivered R1 billion in headline earnings in the first half of this year? (Never mind that it’s down 9% year-on-year!).
Overall, Absa Group’s Retail Banking South Africa division generated R3 billion in headline earnings for the six months, nearly 40% of the group’s total. Based on disclosure, at least, Absa thinks this not material. Investors should take note.
* Hilton Tarrant works at YFM. He can still be contacted at firstname.lastname@example.org.