Why I spent a year counting every bank in regional Australia
Updated: Apr 30, 2022
I have spent more than a year counting banks.
It’s not an introduction to a story I could have foreseen myself writing 12 months earlier but when News Corp started cutting jobs in the rural mastheads, I found myself with time on my hands and a lot of unfinished business.
Every time I wrote about another bank closure, my editor would ask me simple questions I couldn’t answer because either the information wasn’t statistically available or the banks wouldn’t hand it over.
I found the lack of transparency around the sector unacceptable.
The “big four” (ANZ, Commonwealth, NAB and Westpac) started pulling out of regional Australia more than 30 years ago and one of the obstacles to reporting on this has been that there still isn’t a clear picture of the scale of closures.
The “Authorised deposit-taking institutions' points of presence” database of bank locations was set up in 2001 in response to criticism from regional banking inquiries that if this sort of information was not readily available, it was near-impossible to monitor bank behaviour in the bush.
While the Australian Prudential Regulation Authority (APRA) was none too keen to take the job on (it did not consider the collection and analysis of data on the availability of banking services as part of its functions), it was directed to work with the Australian Bureau of Statistics to come up with a survey that captured all service channels available to customers.
The problem is that the system that was developed has never properly nailed the brief to hold the banks to account.
All APRA data reveals is:
How many banks a town has but not which institution owns them; or
How many branches a bank has in an Australian Bureau of Statistics geographical remoteness category, but not in which towns.
Each set of data is displayed in separate tables that can’t be cross-referenced to find out where the different banks have closed branches, resulting in most obvious questions about a bank’s regional footprint being left unanswered.
While APRA has become the official source of information on the banking network in Australia, even a superficial scan of its data has revealed the sort of errors that distort statistics – the same statistics released by government and used by researchers and journalists writing about the state of regional banking service levels.
It’s just a shame that only a redundant journalist has had the time to pick up the errors.
In the defence of anyone who feels they might need defending, it’s not an easy process to define Australia’s regional banking network (open and closed).
The elements of difficulty include:
Where to start?
How to handle mergers and acquisitions?
How to define regional Australia?
The starting point for my records is 1975. There are two reasons for this. In terms of branch numbers, the mid-seventies was the peak of the banking footprint in Australia. Banks were more interested in opening branches than closing them until then so any record of locations from this period would be a definitive list.
The only record of locations I have been able to find from this period is a final print edition of The Insurance and Banking Record from December 1975, which I have used as my first historical reference point.
With the corporate identities of banks changing over the years due to mergers and acquisitions, the banks of this period that have gone on to be part of the “big four” have been included.
To sort regional from metropolitan where the lines blur on city fringes I have used local government definitions of urban and rural.
From this list, I pinpointed closures using BSB records, private research and media reports.
After completing a record of the “big four” I could see that if I wanted to properly represent bank service levels in the bush I would need to keep going and document the minor corporates, mutual banks and franchises.
It was this process that made me realise modern banking had slipped the leash.
The road to APRA
Australian banks published their branch and agency locations in The Insurance and Banking Record from the late 1800s to the mid-1970s. Branches were listed as either city or country, with the decision of where the dividing line fell being an arbitrary one made by the banks.
With the introduction of a numerical identification system administered by the Australian Payments and Clearing Association (now the Australian Payments Network) in the early 1970s, branch locations were each given a code that signified which bank owned the branch, the state in which it was located and a unique number for the branch itself. These “BSB” numbers were published annually in print until 2004 when the records were digitised.
With every bank branch having a unique ID code, it was a relatively easy process to keep track of where bank branches were operating, but in the late 1980s something started happening that hadn’t been seen before – the “big four” had begun to close branches.
Focus shifted from where a bank had branches open to how many towns it had abandoned.
At first the banks were relatively open about their decisions, even announcing closures through the media, but after the excoriation they received when coverage of the sheer number of branches being shut fuelled community outrage, a different strategy was adopted.
Now only those who need to know are told if a town is to lose its bank.
This exclusive list comprises customers, organisations the banks are required to notify under law (such as the Financial Services Union) and the regulatory bodies. By doing this, many closures after the first wave slipped by without media attention and are now difficult to identify.
To their credit, ANZ and NAB have always identified closed branches in BSB listings but Commonwealth and Westpac remove the name of the town and roll its BSB number into the next closest branch resulting in the information being lost. This was not such an issue when print editions were available and previous issues could be referred back to, but digital lists are overwritten and previous versions are not easily accessed.
Despite these challenges, the BSB system is still the most useful way of keeping track of the branch networks – but this is changing.
The digitisation of banking has meant a requirement under the Cheques Act (1986) for a bank branch to be identified by a BSB number is no longer relevant and newer banks are using a universal number for all branches.
With NAB recently beginning the practice of discontinuing BSB numbers when it closes a branch, the BSB system is no longer the tool it once was for identifying which banks are left in a town.
The APRA system is no substitute because it does not reveal this information to the general public. The only other official channel to find out where a bank has – or had – branches located is to get the information from the banks themselves and this is a subject the bigger and older institutions are very coy about.
Full lists of branch locations are either not published or are very difficult to find, and search engines on websites only allow for a small selection of geographically close branches to be revealed at a time.
While the newer banks’ footprints can’t be tracked using BSB or APRA data, they are at least still in that stage the “big four” were in the 1970s and proud enough of their networks to publish the information on their websites for all to see, although I was sin-binned once by Heritage Bank for “suspicious behaviour”.
Unfortunately, until banks are forced to start revealing branch closures to the general public it will take an enormous effort to keep the lists I have come up with updated, which is a ridiculous situation when access to cash services is considered a fundamental right under legislation and banks an essential service.
The addition of the lesser players completed the picture of banking in regional areas and with the help of mapping software I was now able to see which institutions were still providing services in particular locations.
It was at this point that curiosity kicked in.
I wondered what would happen if I loaded the 2019 APRA coordinates into my own spread sheet and mapped them.
The first anomaly I found was the Rural Bank, which I had not included in my records for numerous reasons.
For one, the bank as Elders Rural Bank and later Rural Bank Ltd has been incorrectly classified as having “branches’ since 2001 despite not ever meeting APRA’s own criteria for that service channel, which is to “accept cash and other deposits (including business deposits) and provide change”. (The service, if required locally, was outsourced to Australia Post, meaning the more appropriate APRA classification would be “other face-to-face”.)
With the Rural Bank and Elders having a relationship since the bank was established in 2000, it is not surprising that when the Rural Bank data is dropped into mapping software it reveals the locations of more than 200 Elders stock and station agencies around Australia.
However, the Rural Bank ceased to exist as a separate banking entity on May 31, 2019, and is now a division of the Bendigo and Adelaide Bank network, which according to its website, has “eliminated the costly overheads of establishing and maintaining (Rural Bank) branch offices in rural towns”. Despite this, Elders sites are still being listed as banks in 200 rural locations across Australia.
Why is this important?
Because it skews the figures.
For example, according to APRA data, there are nine bank “branches” in Hamilton in Western Victoria. When I cross-referenced this against my records, I could only find five: ANZ, Commonwealth, NAB, Westpac and Bendigo Bank. On closer inspection I found that APRA had listed a Rural Bank at Elders on the southern outskirts of the town, as well as Rabobank, which does not provide retail services, twice in the one location. There was also a second Westpac branch that was actually a Bank of Melbourne branch that closed some time previously.
As mentioned in my main article ‘Big four cast a dangerous shadow’, Hamilton only had just over half the banks it was supposed to according to APRA when I checked, an error that becomes particularly significant when a town gets down to its last bank, like Casterton did a couple of years ago.
Casterton is a complicated story when it comes to APRA figures.
Sometime between 2014 and 2017, it dropped off the data summary of services by town/suburb and is now only included in a geographical area. (This means it no longer comes up in a name search.) But looking at the 2018 data in map format, it is possible to see that even though Casterton lost its last bank in 2018, APRA continued to show that it still had access to banking services through one last institution – The Rural Bank, which, as stated, does not offer branch-standard service.
Just to make the situation even more confusing, most of the former Rural Bank locations remained in the 2019 data as Bendigo Bank sites, but Casterton, where the Rural Bank would have stood out like a sore thumb in a town bereft of all banking services, is inexplicably no longer there.
The Rabobank errors identified in Hamilton are repeated nationwide, as is the listing of mortgage brokers that have links to Westpac.
Factor in random checks finding a bank supposedly located in the middle of the Maryland National Park and numerous other examples of incorrect classifications being applied, it is obvious the story APRA data is telling about regional Australians’ access to banking services is not correct – and hasn’t been for a very long time.
Checking official data for accuracy is something that takes the sort of time journalists on mainstream payrolls don’t have the luxury of, but they shouldn’t have to.
I did not set out to try and find errors – it is an unintended consequence of following a question down a rabbit hole.
That question wasn’t – as it might appear from what I have written so far – about how many banks are left in regional Australia, but I needed to find that out to get to the heart of what has been playing on my mind when reporting on branch closures over the years.
You see, I was a bank kid.
I grew up living behind the branches my dad managed across regional Victoria. Our friends were mostly bankers. Many of my family members were bankers.
My memories are littered with stories and incidents that have ingrained in me that a bank is a sacred place where cash and valuables are kept. My dad still recalls the smell of the money the Italian fruit growers used to bring in, which he was sure they had been putting in their shoes for safekeeping. And for me, etched in my memory is the Sunday afternoon when a local prospector turned up at home one with a 20 ounce gold nugget he had fished out of the Ovens River to put in a safety deposit box.
From my own experience when reporting on the closure of banks in towns, the most pressing issues for customers relate to cash management: how will they bank their business takings, where will they get their cash floats, how will they access their personal money?
Digital technology has changed the way we bank but while money is still being printed, it is our legal currency and people, particularly those in rural and regional Australia where telecommunications are notoriously unreliable, continue to use it.
With more cash in circulation than ever, the question I have not been prepared to leave lying on the side of the road as the news cycle pounds on is this:
If there is no bank, what is happening to the money?
To find out the answer to this, read 'Big Four' casting a dangerous shadow
In memory of Roly Webster: February 5, 1933 — June 17, 2021