Plus: Charles Schwab, Equity Research, Railsr
Welcome to another issue of Net Interest, my newsletter on financial sector themes. This week, we look at Visa as a new CEO gets ready to take the helm – the fourth in the company’s history as a public company. Visa is one of the biggest companies in finance, but is there any growth left for the new head to deliver? Paying subscribers get access to all my underlying data as well as comments on Charles Schwab, equity research and bank-as-a-service company Railsr. To join them and unlock all that extra content, please sign up here:
Fifteen years ago, I sat down in a coffee shop near my office in Mayfair and opened up a securities prospectus that had recently landed on my desk. Visa had been founded many years before as an electronic payments network but for most of that time it had been structured as a mutual organisation, owned by the financial institutions it served. By October 2007, these financial institutions were ready to sell; the company reorganised into a joint stock corporation and filed for IPO.
The prospectus explained the company well. It described how as a network, Visa sits between bank card issuers and merchants to facilitate payments. After a consumer presents their card to a merchant, Visa routes an authorization request to the consumer’s bank. Provided the authorization checks out, Visa then also routes a clearing message, which contains all the relevant transaction data, and directs the bank to settle funds in an amount it calculates, net of fees.
Visa doesn’t move money – that’s the job of the banks – it moves information. The company gets paid a very small slice of the “interchange” fee that the financial system carves out for itself for intermediating this process. In the financial year before the IPO, that was 0.23% of its payment volume.
At the time, Visa was already big. Its cards were accepted in 170 countries around the world and it was processing 117 million transactions per day. Its growth was underpinned by the ongoing shift from paper-based payments such as cash and cheques to card-based and other electronic payments. In the six years prior to IPO, global card purchase transactions had grown by 14% per year; in the next six, they were projected to grow at 11%.
Leading the company through this was Joseph Saunders, who had been poached away from one of Visa’s member banks to manage the IPO process. He laid out a strategy to expand the network including into new and high growth geographies and market segments. Supported by the tailwind of cash-to-card conversion, he projected that the strategy would deliver annual revenue growth of 11-15%, an operating margin in the low 40% range, and earnings-per-share growth of 20% or more.
Although he was more reticent about it, Saunders also sought to capture efficiency gains to meet his targets. As a mutual organisation, owned by its own customers, Visa had not been run with a strong profit motive. After its reorganisation, the company was free to raise prices and squeeze out a higher operating margin. Revenues inched up to 0.28% of payment volume and the operating margin improved to 63% from 47% in the year before the IPO. Saunders also invested heavily in the brand, spending close to $1 billion a year (equivalent to 14% of revenues) in his first few years in charge to promote the Visa mark around the world.
His job done, Saunders stepped down in November 2012. Distracted for a while by the financial crisis, and the Durbin Amendment which capped fees on debit cards, he was nevertheless able to deliver on all his targets. On his watch, revenues grew by 14% a year and earnings-per-share grew by 33%. The stock price rose in parallel, up almost 30% per year over the period.
Since then, two further CEOs have taken the helm at Visa. Charlie Scharf succeeded Saunders and ran the company for four years (Scharf is now CEO of Wells Fargo). His crowning achievement was reuniting the global company: When Visa reorganised ahead of its IPO in 2008, Europe decided to cling to its mutual roots. An option agreement was put in place to allow Visa Europe to sell to Visa Inc if its members ever changed their minds and, in November 2015, they did. Scharf oversaw the transaction, which promised to deliver $200 million of cost savings – equivalent to 30% of Visa Europe’s operating expenses – as infrastructure was moved to a common platform.
Scharf also had to adapt to the rise of mobile. “Consumers are transacting in large numbers over wireline Internet connections,” said his Global Head of Innovation and Strategic Partnerships at the company’s investor presentation in 2013. “But with mobile devices, we’re just at the early stages of mobile remote payments, the ability for consumers to use the browser on the phone to make purchases… We need to make that easy.”
That year, Visa developed a tokenization service that secured cards on devices by replacing the 16-digit primary account number traditionally printed on the card with a proxy card number that would be of little use in the hands of a fraudster. By driving higher authorisation and lower fraud rates, it has proven to be a very useful innovation. The development effort required over 1,000 developers, consuming the company’s development capacity for almost a year. In its most recent financial year, Visa crossed 4.8 billion tokens, surpassing the number of card credentials. “We think one day, almost every transaction will be tokenized,” its CFO told investors.
Without the immediate benefits of the demutualisation, earnings-per-share growth under Scharf slowed to 11%. But seeing the resilience of the model, investors rewarded the company with stock price returns of close to 20% a year. By the time Al Kelly took over as CEO in December 2016, Visa was doing over $10 billion of annual pre-tax profit, up four times from its IPO.
Al Kelly resolved to continue expanding the network. The growth of e-commerce helped. “As I go back and look over the numbers,” he said in 2019, e-commerce “grows between three and four times face-to-face… And then for us, we pick up 2x plus the share in the e-commerce world that we do in the face-to-face [world] because cash isn’t an option.”
Kelly’s big challenge was the rise of fintech and in particular the threat posed by digital wallets and Buy Now Pay Later (and crypto). I introduced my own briefing of Buy Now Pay Later over two years ago with a question about the credit card industry: “Is that revenue safe?”
It turns out that although at the margin Buy Now Pay Later may threaten credit, it relies very heavily on the debit card rails that Visa operates: Around four-fifths of Buy Now Pay Later instalments are repaid via debit. Because debit fees include a fixed component, splitting a single payment into three or four instalments means more money for Visa. In addition, Buy Now Pay Later providers are increasingly using virtual cards to settle with merchants. A virtual card is a single-use primary account number that incorporates all the security features of a regular card on the network. By using virtual cards, a Buy Now Pay Later provider such as Klarna is able to extend its reach from the 450,000 merchants it directly partners with to the 70 million that accept Visa.
“We’re an enabler for the different ways to pay.”
“We’re an enabler for the different ways to pay,” Kelly told investors last year. “Those different ways to pay don’t compete with us. We enable them to be able to operate in the marketplace, whether it’s crypto or Buy Now, Pay Later … We’re in the business of enablement. We’re not in the business of picking winners and losers. So we’ll enable anything on our network.”
The same is true of wallets. When PayPal launched its wallet, it encouraged users to stay within its closed loop when transacting with merchants. But customers pushed back, and in 2016 Visa (and Mastercard) were incorporated inside the wallet as payment instruments. In addition, as with Buy Now Pay Later, customers use cards to deposit funds in their wallet. In PayPal’s case, around 55% of the wallet is funded via cards.
Kelly responded to the challenges by ramping up investment. One of the features of Visa’s model is that it throws off a lot of free cash. Kelly squeezed out just one percentage point of operating margin improvement on his watch in spite of the increased scale the company achieved over the period (transactions per day increasing to 690 million from 380 million) as he invested back in the business. His M&A strategy backed up his internal investment programme; Kelly spent $3.3 billion on acquisitions, almost 1.5 times what the previous two CEOs spent outside of the Visa Europe integration. Many of these investments were made to support a broader “network of networks” strategy that recognises the more complex payments landscape prevalent today – more on this later.
Next week, Al Kelly leaves his role as CEO to become executive chairman. During his tenure, revenue growth remained stable at 11% a year but earnings-per-share growth picked up to 19% as Kelly used surplus free cash flow to buy back stock and reduce his share count. The stock tracked the growth in earnings-per-share, up 19% per year during his time in charge.
Kelly is being replaced by Ryan McInerney, who was recruited into Visa in 2013 by his former boss at JPMorgan, Charlie Scharf. By all accounts, he is well respected in the company and in the industry. McInerney assumes the mantle after Visa has regained momentum lost during the pandemic. After lagging for several years, cross-border volumes have now broadly recovered. On their latest earnings call this week, company management stated that cross-border volumes are now at 132% the level they were at in the last three months of 2019, with travel-related cross-border volumes at 118% of the level they were at back then.
In addition, the cash-to-card tailwind is flagging. According to the Worldpay Global payments Report, cash use has fallen in store transactions by around four percentage points a year over the past four years, bringing it down to around 18% in 2021. Over the next four years, Worldpay forecasts a shift of only two percentage points a year, to 10% of in-store payment volume globally by 2025.
McInerney will still benefit from growth in Personal Consumption Expenditures (and from inflation) but to enhance that, he will need to look outside traditional consumer card payments. Enter the “network of networks”: “Our network of networks strategy is about moving money to all endpoints and all form factors around the world,” he said in 2020.
McInerney is able to build off Visa’s strong competitive advantage – in essence, that it’s irreplaceable. If you want to build a network to compete, you’d have to win over the banks; except you wouldn’t get the banks unless you have the merchants and you wouldn’t get the merchants unless you have the consumers. Two-sided platforms are hard enough to start cold; payments networks introduce a third set of stakeholders.
Industry expert, Tom Noyes, lists seven core functions that pretenders would have to replicate if they want to compete:
An economic model that encourages shared investment
Strong standardised contracts
Certiﬁcation with active enforcement
Trust – Bank (issuer) role in managing transaction risk
Active customer and merchant support (enabled by economic model)
Ubiquity – it works everywhere all the time
Innovation – Shared investment, standards and integration
In the definitive history of Visa, author Dave Stearns argues that rather than computer systems and telecommunications underpinning the network, it is the operating regulations that allow participants to coordinate that make it work. “The operating regulations establish the ‘rules of the game,’ and as such, created a common set of expectations and incentives amongst member banks, merchants, and cardholders. These expectations and incentives helped not only to coordinate inter-organizational work, but also to balance out the conflicting interests of these various participants, many of whom are competitors of one another.” One way of thinking about Visa’s edge is that you could never get so many parties to agree.
Yet governments are pretty good at laying down rules too and many of them are sponsoring their own real-time payment (RTP) systems to compete with Visa. We discussed the growth of Brazil’s Pix in Paying Faster back in April. Visa recognises the risk. Its regulatory filings contain a risk statement warning that “Government actions or initiatives such as…real time payment initiatives by governments such as the US Federal Reserve’s FedNow or the Central Bank of Brazil’s Pix system may provide competitors with increased opportunities to derive competitive advantages from these business models, and may create new competitors, including in some cases the government itself.”
But real-time payment (RTP) systems have been around for a while and in most countries they have failed to dislodge the card networks. In Japan, the first country to introduce one, real-time payments represented just 10% of electronic payment transactions in 2020 (around 3% of all transactions). And in the UK, where faster payments really are very fast (as I wrote in Paying Faster) real-time payments represented 8% of electronic payment transactions in 2020 (around 6% of all transactions).
“Comparing an RTP transaction in 2022 to a Visa Debit transaction in 2022…is kind of like comparing a rotary phone…to an iPhone.”
“In general, we haven’t seen, in most countries around the world, real penetration in the kind of the core point-of-sale C2B [consumer-to-business] spaces,” says Ryan McInerney. “And there’s a number of reasons why … I like to use the analogy that comparing an RTP transaction in 2022 to a Visa Debit transaction in 2022, let alone a credit transaction, is kind of like comparing a rotary phone, like a landline rotary phone, to an iPhone. I mean, you can talk in real time to another person on both of them, but the comparisons end there. An RTP transaction today is instant, it’s permanent and it’s irrevocable. A Visa Debit transaction…We have disputes. We have chargebacks. We have globally accepted and adopted operating regulations and rules. We have global acceptance. We have zero liability in fraud protection. We have tokenization. We have a broad suite of risk management fraud reduction tools. I mean, the list goes on and on.”
In the meantime, Visa is able to benefit from increased adoption of electronic payments that real-time payments systems foster. In its most recent quarter, Visa reported “strong double digit growth” of cards in India and Brazil, where the most advanced RTP systems operate. It has also invested in infrastructure to gain access to the real time account-to-account (A2A) market. Visa’s acquisition of Earthport in 2020 gives it direct connections to the local payment systems of 88 countries; its acquisition of Tink in 2022 gives it exposure to open banking in Europe.
At the same time, Visa has adapted its core functionality to compete. As the company’s CFO explained recently: “Five or six years ago, we made some changes to our network to allow the money to move both ways. So it can not only go from you to someone else, but it can also come to you.” In its most recent quarter, this Visa Direct platform did 1.9 billion transactions, up 39% year-over-year – mostly person-to-person payments but also gig economy payouts, remittances and insurance payments. It’s part of a broader strategy to tap into new money flows outside the traditional card use case. By integrating its card network to alternative systems, Visa is positioning itself as a global payments router. The company reckons that together with Earthport, Visa Direct gives it access to a $65 trillion market of lower value, higher velocity flows. Its Business-to-Business strategy allows it to access a further $120 trillion of higher value, lower velocity flows.
As well as going after new money flows, Visa is looking to enhance revenues by providing a range of “value added services” in addition to its core payments offering. These services encompass cybersecurity, fraud analytics, loyalty and more. They stem from the research Visa has conducted over its vast pool of transaction data over many years. Historically, these services were tied to the card business but they are now being sold separately including on competing rails. For example, Token ID applies tokenization to real-time payment systems and CyberSource, which Visa acquired in 2010, is used by bank clients looking to compete in the acquiring business. In its most recent quarter, revenues from value-added services contributed over a fifth of total revenues, and grew at 20% over the year, faster than the overall rate of revenue growth.
Visa’s numbers are big. It is active in 200 countries; operates 3.6 billion cards; is accepted at 70 million merchant locations; processes 242 billion transactions a year, representing $11.7 trillion; and generates over $40 billion in revenue on which it earns one of the highest margins in the S&P 500. Which makes growth hard. It’s a different story from the one laid out in the IPO prospectus, but it’s no less exciting. Let’s see what Ryan McInerney does with it.
Net Interest is a reader-supported publication. To receive new posts and support my work, consider becoming a paid subscriber.
For those interested in a longer-term history of Visa, my piece, Dee Hock, the Father of Fintech, may be useful.
Kelly inherited Visa Checkout, a digital payment service the company launched in 2014. It simplified the process of entering the long card number at the e-commerce checkout, allowing users to enter a username and password instead. After a few years, though, he dropped it. Rather than fight for market share against all the other checkout buttons, his preference was for an industry standard and he backed the creation of Secure Remote Commerce (SRC) which brings up the user’s wallet allowing them to choose which card to use in a tokenized format.