February 16, 2020 - 11:59pm EST by
2020 2021
Price: 135.00 EPS 9.00 09.90
Shares Out. (in M): 808 P/E 015 0
Market Cap (in $M): 109 P/FCF 015 0
Net Debt (in $M): 0 EBIT 0 0
TEV (in $M): 109 TEV/EBIT 0 0

Sign up for free guest access to view investment idea with a 45 days delay.

  • Underfollowed
  • Misunderstood Business Model


American Express is a business that can serve as a foundational brick for a long term portfolio.


Sometimes our biggest error is getting in our own way, impeding the path of compounding. At approximately every moment in its 170 year history, AmEx has been a good buy. Today, it remains a business growing secular mid- to high-single digits both top and bottom line. It requires little incremental capital to drive that growth, allowing it to return over 80% of its earnings to owners, which it does religiously.


AmEx is effectively a mix of MasterCard or Visa and a premium version of Capital One. Only 20% of American Express revenue is from the lending side of the house, while 80% is derived from fees (primarily swipe economics and membership fees). This attractive 80:20 mix has remained fairly stable for many years. 


While Capital One and its bank brethren trade at low double digit P/Es, Visa trades over 35x 2020 (and faster growing MasterCard at 38x). The S&P 500 trades at 20x 2020 earnings, and has slower expected growth and greater capital needs to support that growth than AmEx. 


Visa’s premium valuation may well be warranted. I grant that Visa is one of the world’s great businesses. It’s just that it turns out AmEx is not too shabby itself. Compared to traditional banks, American Express has far superior business quality. It’s 30%+ ROE highlights this (vs. Capital One at 10%-12%). 


American Express is a remarkable business with a differentiated franchise and a secular growth runway. It is not a sprinter. It’s a world class distance runner, designed to persist, outlast, and win the long-term.


AmEx is available for 15x 2020 earnings. Attractive absolutely and cheap relatively.





This is a story about a company that does the work for us. We don’t have to bet on management genius; we simply need them to keep executing, investing in their moat, and sending us our profits. The rest will take care of itself. 


American Express isn’t sexy. Rather, it is foundational. It is the kind of company one can buy and put at the front of the portfolio train, helping tug returns in the right direction and liberating you to focus on filling the cargo cars with valuable freight. 


At its simplest, this is a business growing net income at a sustainable 7% CAGR and available at a 7% earnings yield (most of which is distributable). That simplified 14% return algorithm gives no credit for change in multiple. 


Given Visa is a better business and quite popular amongst savvy investors, one might think its return algorithm would be far superior. Visa trades at a <3% earnings yield and is growing 11-12% p.a., also providing a 14% expected return. [Note: I'm taking a longer term view of net income growth, not a one year expected number, which would be higher for both companies.]


Both may be attractive. I happen to prefer AmEx: A) its return is made more robust by the larger weight on current cash flow; and B) given its above-market growth but below-market multiple, it has more positive re-rating optionality.  In my opinion, it is much more likely we see AmEx trade for 20x than Visa trade for 50x.


American Express sits in broad daylight as an unsexy, robust, mid-teens or higher compounder.  Sometimes our job really is to get out of our own way.





AmEx is Robust:

I can write endlessly about the business.  About its mistakes in the lead up to the financial crisis and its lessons employed since then. About the quality of its lending book, funding sources, and reserve policy. About its differentiated and very hard to replicate membership benefits. About its ability to bounce back from losing CostCo four years ago. About it powering straight through, as if unscathed, extremely aggressive challenges of AmEx’s premium consumer position (e.g. Chase Sapphire). About its absolutely dominant position in commercial (your corporate AmEx). About its giant remaining opportunities in US SMB and international.  


But the reality is, all the depth of writing in the world does not prove it nearly as well as AmEx’s objective performance. American Express has grown its topline at GDP+ since time immemorial. Over the years, it has diversified its business from an overwhelming reliance on T&E (travel and entertainment spend), which is more economically sensitive, to a broad-based corporate spending partner, a premium consumer brand, and an enviable lending franchise. 


While T&E is still an important and highly differentiated part of its business, AmEx has expanded its underlying drivers.


A Straight Forward Business Model:

At its core, the American Express business model is fairly simple: target high spend clients, issue them cards, charge fees for premium cards, collect swipe fees, and lend where appropriate. 


I think you can conceptualize the revenue build in a fairly straight forward manner:


As we look at the core drivers above, AmEx sits on top of massive global trends. It takes a vig on the plastic portion of global GDP. That portion is growing faster than GDP because plastic (in both physical and digital form) is taking share from non-plastic, and AmEx is holding its share in the US and taking share (from a very small base) abroad.  


Importantly, AmEx holds a special place in the plastic ecosystem, as the premium industry brand and a leading issuer, network owner, and processor. Since AmEx generally does not have to share swipe revenue with card issuing partners (e.g., JPM or Capital One) or networks (e.g., Visa), and its customers are very large spenders, it can afford to provide differentiated rewards, service, and experiences that its competitors struggle to replicate. This creates stickiness and ongoing subscription-like economics that it can reinvest in the customer proposition and acquisition.



I suspect most readers of this write-up know that AmEx has gone further than just that. I bet most of you have an AmEx - likely Platinum - under your ass or in your purse, as you read this. You likely know that your Platinum card conveys a suite of benefits that are quite remarkable, not least of which are things like your monthly Uber credits and, importantly, access to proprietary AmEx and partner airport lounges all over the world. You also collect premium spending points, among many, many other benefits (e.g. AmEx pays for your TSA Pre-check/Clear fees, you are automatically given gold status in the Hilton system, premium hotel reservation upgrades at many of the best hoteliers in the world, priority concert ticket purchases, etc.). 


Annually, American Express is spending close to $20 billion on Customer Engagement. 


Read that again. 


Over 60% of that spend is on Rewards and Card Member Services. In particular, its unique offerings - which show up in Card Member Services - has been a rapidly growing expense item.


The balance of that $20 billion is spent on marketing and business development (BD includes payments to partners, like Delta Airlines, 3rd party AmEx issuer payments, etc.). As AmEx enters long term renewals with partners like Delta or Hilton (you may have a co-branded Delta AmEx for instance), this is driving up Business Development expense. 


BD and investment in Rewards and Card Member Services have replaced marketing spend, which has remained largely flat for years. Likewise, as AmEx has automated more and more of its cost structure (think digital customer service in its app vs. receiving phone calls), it has freed up margin for other uses. In essence, AmEx has repurposed former G&A and marketing spend to drive value to its customers and partners, which has led to higher quality, stickier customers. It is scale begetting scale.


In exchange for this massive investment in returning value to its card holders, AmEx charges as much as $600/year in membership fees (for Platinum cards). It also has gold cards and centurion (green) cards, that charge lower fees for various levels of benefits, as well as non-fee cards. In 2019, of the 11.5 million newly issued AmEx’s, ~70% were fee-based cards.


With $1.2 trillion in annual total billed business (i.e., spending on AmEx), it is - far and away - the largest issuer of premium cards. This investment in premium customers and high spending activities (e.g., travel), has given AmEx a special place in the world. While its number of cards in force is much lower than Visa or MasterCard, its spending per card is on another planet.  The average Visa/MC customer spends around $5,500/year on their credit card.  The average proprietary AmEx customer spends $20,000/year. This high spending nature makes AmEx customers extremely desirable, not just to AmEx, but to merchants and its partners.


Between the cardholder fees and its swipe economics, AmEx has an unmatched annual subscription-like war chest with which it can invest back into cardholder rewards, in particular fixed-cost rewards and experiences. It is one thing to give out points to cardholders based on variable spending. It’s another thing to make firm commitments to spending of billions of dollars annually on partnerships, services, and rewards. 


In order to have customers that spend so much, those premium and corporate customers need to avoid frequently bumping into spending thresholds. AmEx is known for its very high and flexible spending limits. Rather than setting a fixed threshold, for many of its premium customers, it instead uses its more nuanced understanding of its customers and its advanced technology to provide variable limits, and it can override those to provide additional spending power when needed. You cannot be taking clients out to dinner and have your card be rejected because you hit a ceiling. AmEx gets this.


Behind that approach to spending limits is something I’ve not spent much time talking about, and don’t plan to in this write-up, which is American Express’s advantage on information. Because AmEx is both the issuer and the network, it has a “closed loop” picture of its customers’ behavior. This allows AmEx to differentiate on a variety of fronts, from marketing, to anti-fraud, to lending. Importantly, it allows the company to gather more data on you and provide you with much higher credit thresholds than most traditional issuers.


The result of all of this is that while AmEx is smaller than Visa and MasterCard, it is not “small”. It has 93 million cards-in-force, of which 43 million are domestic. Its network is effectively tied with MasterCard as the second largest credit card processing network (by dollars) in the US. [If debit is included, American Express falls behind]


Likewise, on the issuance side, AmEx is approximately tied with Capital One (behind Citi and Chase) as the third largest credit card issuer, by number of cards. 


Ranked by spending, American Express is by far the largest issuer.


This is battleship AmEx, and it continues to steam ahead.


Owning a Two-Sided Market:

This combination of fee revenue, cardholder relevance, and massive in-place scale create significant barriers to entry in the premium and corporate space. But they only represent one half of a two-sided marketplace.


The other half is merchant acceptance. Few things have been more frustrating to AmEx cardholders than the phrase, “I’m sorry, but we don’t take American Express.” 


For many years, that was a perfectly rational response for many small merchants. AmEx charged premium swipe fees for the privilege of a merchant having American Express cardholders spending their money at your store. This left American Express vulnerable and effectively forced its otherwise loyal cardholders to carry (and use!) a second card in their wallet. Over the past decade, and accelerating over the past five years, AmEx management made closing the acceptance gap in the US a strategic priority. This has required American Express to eliminate its premium swipe economics from most merchants (it still charges a premium in the T&E industry, where merchants need to accept American Express to serve business customers). As the cost of accepting AmEx has fallen (and the cost of taking Visa and MasterCard has actually risen), merchants have decided to take them all.


While there are still many merchants that do not accept American Express, the number today is quite small. As of year end 2019, 99% of credit-card accepting merchants in the US are now able to accept an American Express. 



International remains a wide open, and fast growing, opportunity set for American Express. 


Its focus on proprietary issuance is in a narrow, targeted set of countries. It further participates in issuance by playing a more Visa and MasterCard like role in non-core countries via its modest GNS business (which allows for third party AmEx issuance). 


In those core issuance countries, American Express of course focuses on broadening acceptance. However, it also works to grow acceptance outside those areas by “following its customers.” If its customers show a propensity to spend in certain areas of Iceland, AmEx wants to be accepted there. As it grows local acceptance for its global customers, it can begin to focus on growing local issuance, pushing the fly-wheel faster in those markets.


This international expansion is an ongoing strategic priority. In the company’s own words, from the recent Q4 conference call: 

“We're also making good progress to increase coverage across our international markets where our Card Members live, work and travel to the most, and this will continue to be a focus for us. Going forward, as we continue to grow our network, we'll work with our merchant partners in the U.S. and around the world to ensure that our Card Members are warmly welcomed and encouraged to spend in the millions of places where their Amex cards are accepted.


“Finally, I'm also pleased to report that the People's Bank of China officially accepted our network application, an important next step in our plan to build the network business in China.” [note: this would make AmEx the first US player to build a network in China]


AmEx is far from parity coverage in most of these non-US markets. Addressing that and picking up the concomitant growth will effectively be an evergreen opportunity for the company.



Lending is a core business, but unlike AmEx’s card-issuing competitors which are issuing cards as an asset (lending) creation machine, AmEx views lending as yet one more profitable service it can provide its premium customer base. 


Historically, American Express was known for its “charge” cards, which required monthly repayment in full.  On the corporate side, it largely still is a charge-based card model, though increasingly it offers targeted small business lending products as part of its offer. However, the bulk of AmEx’s lending business is consumer based.


The average domestic American Express consumer makes nearly 50% of their plastic purchases on their AmEx, but carries only 20% of their total credit card lending balances on AmEx. That gap is a large and addressable opportunity within its existing customer base, and American Express has been going after it. 


Unlike many other card issuers, who acquire new customers by offering low-interest balance transfers (and therefore potentially pursue an adverse selection business model), American Express is not a heavy participant in that space. Instead, nearly 60% of its loan growth is to customers who have been with American Express for over a year.


Unlike many card issuers, where credit card balances are really a subprime lending product, American Express leans toward unsecured prime and super-prime. It has far and away the best credit performance in the industry.  I can go into more depth on this in the comments section, but loan performance has consistently performed in-line or better than its models, leading to attractive reserve evolution.


Funding this growth is a mix of liability sources, from ABS to term debt. Most important has been deposit funding, which makes up more than 50% of its loan funding. 


As can be seen in the graphic below, the attractive NIM combined with manageable write-offs, has led to a high quality asset franchise that supplements the spending and fee-centric core business. I expect the lending business will continue to grow faster than the rest of the business, though the overall 80:20 mix will not change quickly.



While charge card balances require some amount of capital in order to float customers for a payment cycle, American Express’s real capital requirements are around its lending business. Growth in spending and lending mean equity capital also needs grow with the business in order to stay leverage-neutral. While for the past many years, American Express has been able to return 90-100% of its earnings to shareholders (between dividends and buyback), that was facilitated by coming from a dramatically overcapitalized position to its current 10.5% - 11% Tier 1 common equity position. I expect it to stay in this range.


Every year since the financial crisis, the Federal Reserve has stress tested the largest banks in the US (of which AmEx is technically one) via a process called CCAR. One result of this test is a capital plan that defines how much capital each bank must retain for the coming year and how much it can return to shareholders. American Express has regularly been the best performing bank in the Fed’s stress test process, and the Fed estimates AmEx would be profitable all the way through a new crisis (as it was during the GFC, albeit still taking higher than expected asset losses). 


I’ve included a few relevant slides from last year’s investor day that make the point more effectively than I can.