09 March 2017
Payment Instrument in Profile: The ATM
This article is the first in a series that will examine how we use and experience different payment instruments in the UK. Each report will focus on a specific payment type, such as cash, debit card, direct debit etc., and discuss the key trends and future prospects of the different ways we pay for things.
While many of us may not consider it a payment, withdrawing cash from an ATM is one of our most common transactional activities. With 2017 marking the 50th anniversary of the UK’s first ATM withdrawal, what better place to start than to examine the history and current fortunes of arguably the most important initiative to bring everyday banking to the masses in the history of financial services. OK so that may seem a bit of an exaggeration. But let’s remember, before mobile, internet and telephone banking, the ATM was effectively the world’s first remote banking service that consumers had access to. A late night trip to the ATM may not seem a particularly revolutionary action in our age of near saturation of card acceptance at retailers and banking on our phone, but at a time when the only access to our money was through the restricted hours kept by bank branches, the ATM was ground-breaking. Even today, the humble cash dispensing machine plays a crucial role in bringing access to banking services to millions of people around the world. One of my favourite stories of recent years was the floating bank branches and ATMs of Indonesia that take basic banking services to the country’s remote islands. A recent report by Retail Banking Research reported that ATM cash withdrawals worldwide grew by an impressive 10% in 2015, the fastest rate since 2011. That’s 99 billion transactions. According to the research, financial inclusion is the key driver of this growth, particularly in Asia Pacific.
There is also a rather personal reason for wanting to write about this topic. The first ATM in the UK was installed at a branch of Barclays bank in Enfield Town in 1967. As a kid, my mum would regularly take me on shopping trips to Enfield Town. I have vivid memories of traipsing around Pearsons department store (still there), which is opposite Barclays bank, hoping to get a brief hiatus in the toy department. As we walked past Barclays, it would often be commented that we were standing in the midst of the world’s first ATM. There was a strange sense of civic pride in this fact, after all, it did make the place where we lived a little more interesting. OK, it is not quite up there with Runnymede (Magna Carta) or Stratford upon-Avon (my colleagues tell me a famous writer of sorts used to live there!), but for me, as a young boy, this was a site of genuine historic importance. Later in my “student days”, I and a group of like-minded friends would every Saturday afternoon set up stall outside Enfield’s ATM prime and petition passers-by on various social causes of the day. If there were any new members to our troupe, I would insist on giving them the full guided tour of this important piece of local history.
The arrival of the first ATM in June 1967 had its roots in a more general self-service trend that began to take shape in the UK in the 1950s and 1960s. With the end of war rationing, which lasted way into 1950s, a whole new consumer consciousness began to emerge. As part of this, the UK looked to the explosion of consumerism in the US and, in particular, the increased convenience of self-service. This was a time of the first self-service supermarkets and petrol stations, which were seen as more efficient for both customer and retailer, transport ticketing machines and vending machines. Indeed, the idea for the ATM, which Barclays dubbed the “robot cashier”, came from John Shepherd-Barron of De La Rue Instruments whose eureka moment occurred when he arrived a minute after his branch had closed one Saturday afternoon and decided enough is enough. His inspiration was the popularity of self-service chocolate vending machines.
Although I should set the record straight, Barclays’ ATM was the not in fact the world’s first. The first operational cash dispensing machine was installed the previous year in Japan. However, for the purpose of this article and in keeping with the new post-truth era we are now apparently living in, I’m going to conveniently skip over that fact for the sake of my younger self’s happy delusion.
According to Barclays, the newly launched cash dispensing service, known as Barclaycash, was designed to dispense £10 against a special paper voucher which the customer inserted into the machine. By keying in a personal code number (initially 6 digits rather than 4), the customer could obtain cash in a matter of seconds.
Card-based ATM cash withdrawals were introduced in the UK by Lloyds Bank in 1972. Known as a CashPoint, the IBM developed machine used the magnetic stripe on the card to identify the customer’s account. The cardholder would then authorise the cash withdrawal using a PIN. As the number of cards in circulation grew, so did the number of ATMs. By the 1980s, bilateral agreements between different banks and building societies allowing users to withdraw cash from ATMs not owned by their bank led to the creation of several shared ATM networks, including LINK. By the end of the decade, LINK emerged as the single entity responsible for the UK’s shared cash machine network as rival networks merged with it and its membership increased until all UK ATM owners had joined the scheme.
The significance of the LINK scheme’s impact on the growth of the ATM market is twofold: 1) Its business model created the market conditions for the strong development of non-bank independent ATM deployers (IADs), and 2) a vast, mostly free to use, network meant cardholders could conveniently withdraw cash from a UK ATM regardless of who owned it.
How LINK’s business works
LINK is based on an interchange model. This means that when a customer of bank A uses their card in the ATM of bank B, bank A pays bank B a small fee for the privilege (there is also an even smaller processing fee that both bank A and bank B pay to VocaLink). This model was originally planned as a cost recovery model based on the cost of allowing banks with ATMs and those without to all share the value of the network equally. As the popularity of ATMs grew, a small number of non-Bank ATMs began to spring up connected to the LINK network. These were initially providers of convenience ATMs whereby they charged the customer directly and therefore did not receive an interchange fee from the issuer. Overtime, the reduction in the cost of operating ATMs for IADs, particularly those who moved previously out-sourced ATM services such as maintenance and cash management in-house, meant that some of these providers were able to create a margin based on the interchange fee.
As more IADs moved to an interchange model, this led to greater usage of their ATMs as they were free to use by the customer, which in turn encouraged IADs to deploy more free-to use ATMs. This virtuous circle, means there is a high penetration of ATMs across the UK, which is great for users as you are never that far away from being able to get access to your cash.
While ‘Pay for’ ATMs are still typically operated by IADs, as the number of these non-bank ATMs has increased in recent years, the actual number of ‘pay for’ ATMs has decreased significantly from a peak of 42 percent in 2007 to just under 25 percent at the end of 2015.
The combination of a widely distributed network of ATMs that are also free to use means consumers are encouraged to withdraw cash on a regular basis, but also, unlike some other markets, are not penalised when they withdraw small amounts. For example, where there is flat fee to make a cash withdrawal, it is cost effective to make fewer and larger cash withdrawals. This is typically not the case in the UK, so, as a result, average cash withdrawal values are comparatively low, while the number of cash withdrawals per capita is very high. In a comparison of EU countries with a population over 15 million, UK consumers make on average considerably more ATM cash withdrawals on a per capita basis (43), compared to the likes of France (27), Germany (19) and Spain (13), while, with the exception of the Netherlands, it also has the lowest average cash withdrawal value.
Declining cash and long term trends for ATMs
One of the big payment trends in the UK over recent years has been the declining use of cash. There were 2.7 billion fewer cash payments between 2013 and 2015, equivalent to 19 percent decline over the period.
There has been considerable noise in the press and among analysts regarding the role of contactless card transactions in reducing cash usage in the UK. While it is undeniable that contactless payments has changed the way we make low value payments, including transactions typically carried out by cash, contactless alone cannot explain this decline. For example, there were just over 1 billion contactless card transactions in 2015 and around a third of that number in 2014. Even if we assume that all these transactions were from cannibalising cash payments, which is highly unlikely as many will simply be replacing PIN-based card payments, this is still around half the number of cash transactions lost since 2013. What’s more, the average contactless transaction (£7.38) is roughly half that of a cash transaction (£14.70) in 2015. This means one typical cash transactions is equivalent to two contactless transactions, which further diminishes its potential impact. Therefore we need to consider a number of other important factors, alongside the rise of contactless, that have contributed to declining cash usage.
The increasing digitalisation of our economy is one such factor. It is not just that we are increasingly buying more online (although this is significant), it is also the fact that how we purchase products and services is changing. The rise of subscriptions-based services, such as Netflix and Spotify, is rapidly affecting how we consume, particularly low value entertainment products. This is having a big impact on how we make payments, increasing our reliance on non-cash payment instruments. Of course this is just one example of many, but the key thing to conclude here is that the fall in cash usage that begun in 2013 is not an aberration, but likely the beginnings of a much larger trend, as new service and distribution models are developed that embrace digital.
Payments UK conservatively forecast that cash would decline by 3.2 percent in 2016. However, having spoken to one of their lead analysts recently - they expect this decline to be significantly revised up when they release their full year figures in Q2 2017. I would strongly agree with this assessment. The latest contactless data for 2016 alone suggests there is likely to have been a significantly higher rate of decline in cash usage last year.
In the long term, the decline in cash is going to have a negative effect on ATM usage. However, while overall ATM cash withdrawals have declined, falling on average 1 percent annually between 2013 and 2015, LINK volumes have managed and continue to heroically buck this trend. LINK accounted for approximately three quarters of cash withdrawals in 2015. Perhaps even more surprisingly for those of us keeping a close eye on LINK volumes, the latest data for 2016 shows a marginal increase on 2015.
The main reason for this continued growth in LINK ATM cash withdrawals in 2016 is that a number of banks sold off their remote ATM estates (i.e. ATMs owned by the banks at non-bank locations such as supermarkets) to IADs. A LINK transaction occurs when you use a card from one bank in an ATM belonging to either another bank or a third party. Hence, fragmentation in the market is good for LINK and can help explain its growth in volumes in contrast to cash trends. However, growth is marginal – so what we are seeing here is an expected decline being temporarily held back by increased fragmentation. Given the size of growth in 2016, just 0.4 percent, it also worth pointing out that 2016 was a leap year. If it wasn’t, there would have likely been a slight decline. When dealing with such small margins, an extra day can make all the difference!
Can LINK continue to buck trends?
With further fragmentation in the market unlikely in the short term, we expect annual LINK volumes will decline in 2017. The sale of bank-owned ATMs in non-bank locations to IADs has been mostly complete and there are no large planned bank break ups, such as when Lloyds split off TSB. That said any decline is likely to be extremely modest. Rather, what we expect is that LINK volumes will remain relatively flat over the coming years, with decline in the longer term. That is not to say that changes in the market environment could not affect this outlook for better or worse. Cash is still a preferred use in a number of payment scenarios. For example, many of us still use cash when going for a night out or taking part in various leisure activities. Increased spending in this category could help boost volumes. Similarly, if the proportion of ‘pay for’ ATMs continues to fall, this could encourage consumers to make more lower value cash withdrawals which could help boost volumes. Whatever the case, as we celebrate the ATM’s 50th year, I can confidently say that the ATM will continue to remain a core element of the payments industry for many years to come.
 Payments UK, UK Payment Statistics 2016