Cardtronics (CATM) manages or owns about 227,000 ATMs, the only pure-play public company in this business. Although ATMs can and do expand their functionality, they are still primarily cash dispensers. If we all believe cash is dying, why is Cardtronics still a good investment?
In a sense, ATMs do not compete with the rapid growth of noncash payments. An ATM transaction is not a payment at all. It is a value mode transference: from paper to centralized record keeping (i.e., the computerized back office) or – more commonly – vice versa.
Why is it called automated teller? Going back in time, tellers were the ones who told the bank customer how much he or she had just deposited, how much the account contained, or how much they could withdraw. They could tell how much loan payment was due or how much interest might have been earned. And they gave paper documentation once the transaction was complete. They were low level; how much a customer could borrow was an officer function and tellers weren’t officers.
Fast forward 100 years. The telling function is now totally electronic. No ATMs are needed for that – although people still use them for that purpose. Put simply, ATMs exist to convert paper to electronics and vice versa. The paper can be a check or cash, it doesn’t matter. If we didn’t live in a long-term twilight era halfway between an all-paper world and an all-electronic world, we wouldn’t have ATMs.
But the fact is we do. Negotiable currency that’s been a mainstay of life for centuries doesn’t die out in just a few decades. People still use it and carry it. The highly banked in the U.S. still get nervous without a single dollar bill in their wallet/purse even if they use a credit or debit card for everything. On the other end of the income spectrum, citizens in underdeveloped rural countries (half the world) use currency as a standard, even if transferring funds from their prepaid mobile balances is growing like wildfire.
Even in the U.S., more currency is in circulation every year. Ditto for euros in the E.U. However, it is not true that more ATMs are in operation every year. The global herd seems to have peaked at about 3.3 million depending on which source you follow. This makes sense. Global population is rising, and so is the share of that population with higher incomes. Global payment transaction volumes are rising even faster than the population, but noncash payments are skyrocketing. Electrons are very, very cheap and eliminating paper from the payment mix leads the emergence of all sorts of new capabilities.
The rate of decline of the global ATM installed base is likely to be very slow. It may even be stagnant for some years – maybe only ATMs per capita will be declining. If ATM volumes are the same, somewhat lower, or even slightly higher in 2029, that would not surprise me. If they were half of today’s number, that would surprise me.
One does not get the feeling that Cardtronics success strategy is based on growth in the number of ATMs. In fact, its average number of transacting ATMs declined by 0.6% in the past 12 months. Of course, the number the company owns or manage may grow through acquisitions or taking over others’ networks. its owned and managed share of installed ATMs is roughly 6–7% globally and about 50% in the U.S., where 60% of its business is.
Of course, there is much more to thinking about ATMs than just how many. For example, cost. If they got a lot cheaper to build, maybe there could be more. If new functions were added, maybe the transaction volumes would grow. Cardtronics’ ATMs average about 25 withdrawals per day, which leaves a lot of unused capacity. The inner-city experience of high-traffic waiting line ATMs generally belongs to the big banks. From Cardtronics’ point of view, higher transaction volumes have high marginal benefits because, basically, the fixed costs are high and the variable cost is very low, close to nonexistent.
Yet, how does one encourage more ATM traffic? In the broadest sense, by adding more functionality. Cardtronics is currently rolling out its own deposit-taking machines – 1,000 scheduled for 2019. This function is not completely new, as the decades-old method of an envelope and deposit slip was enhanced by Check 21 in 2006 to allow image capture and check truncation by the machine. But Cardtronics’ network, Allpoint, is mostly located at retailers, where deposit-taking has never been a factor.
Cardtronics gets only 5% of its revenue from the 139,000 machines it manages for others. This is a repeatable business under longer-term contracts but mainly serves to help cover fixed costs. For the machines it owns, the largest share of revenue (>70%) comes from surcharge fees and interchange. Neither of these sources is growing. Consumers generally hate surcharge fees – which keep increasing. Interchange itself is stable in the U.S., but LINK in the U.K. cut interchange recently, hurting Cardtronics’ economics there and driving Cardtronics (and other independent ATM deployers, or IADs) to introduce surcharges.
Cardtronics’ only growth revenue is from its bank-branding and surcharge-free network, although this was minimal last year. The idea is to piggybank off card issuers who either have no network of their own or who are reducing it (for example, USAA, a major financial institution that operates without brick-and-mortar branches). All such balance holders, including many new fintech players, issue debit cards, but the question is whether they can do better by paying the convenience fee directly to Cardtronics and thereby making their card more attractive. Another opportunity is brick-and-mortar banks or credit unions that either are cutting back their branch footprint or wish to expand in a new service area without investing so heavily in branches.
The idea is that with deposit functions, no surcharge, and new ideas like cardless (about 11,000 such machines so far), Cardtronics can continue to grow its per-machine volumes by signing up more and institutions that believe they can get by with attractive debit cards and no brick and mortar. This is a strategy of deposit disaggregation, as balances run off the established – and generally larger – financial institutions.
How well is this working? The results are still early. Cardtronics is a cash cow and has been so for most of the past five years except for the major disruption caused by losing the 7-Eleven account. That was 12.5% of Cardtronics’ revenue and about 40% of its gross margins in 2017. The stock began to tank in early February 2017 from a high of about 55. By November, it was down to about 18, a possible overreaction. Since then the stock has gradually improved.
For Q1 2019, the company announced $20 million of adjusted free cash flow and also authorization for a share buyback program. These are encouraging signs.
In that same first quarter, across all of its ATMs, Cardtronics saw a 2.3% decrease in cash withdrawal transactions. The number of ATMs that it or retailers own declined by 5%. Cardtronics saw 753 withdrawal transactions per month per machine excluding the ones it manages for others. Adjusted operating gross profit per machine per month went up slightly to $324. So, there is hope for better growth, but it may be slow. In the meantime, rationalization is at work.
Others have noticed. Cardtronics (CATM), currently trading at 34.92, is rated 9.1 “very bullish” from the Thomson Reuters StarMine Equity Summary Score. There may be some takeover possibility built into this price. In 2018, Fiserv bought MoneyPass, the second largest U.S. ATM network, and ancillary assets, from the Elan unit of US Bancorp for a reported $690 million, or about 4 times sales. This compares to Cardtronics’ current price/sales (most recent quarter) of 1.23.
In summary, Cardtronics should remain one of the best value plays in the bank technology space and certainly the only one taking advantage of the very long tail of the ATM existence.