Amex is The Hinterland’s largest holding. It has generated a 25%+ return at the time of writing and it is a long time favourite of Warren Buffet and Bruce Greenwald. However, the question must be asked at least annually, is it a hold?
One can see why so many eminent investors favour this company. It consistently generates gross profit margins of c.50% and returns on equity of c.25%. Its net earnings margin is consistently c.15%, which is not spectacular but very good for a financials company. Plus, over its long history it has built a brand that is among the most trusted across all industries.
Amex is one of the ‘big four’ in the card market, including Visa, Mastercard and Discover. It positions itself at the premium end of the market, targeting wealthy customers, which is a nice place to be. This wealthy customer base creates a virtuous circle because they generate relatively high spend versus competitor cards, which enables it to charge merchants higher amounts per transaction, which enables it to offer customers added value through rewards and benefits, which in turn drives higher spending…and so on.
The other cornerstone of Amex’s competitive advantage is that because it acts as both a card issuer and card network, it creates a “closed-network” when processing payment from an Amex merchant. This is contrast to the “open-network” model in use during a Visa or Mastercard transaction, due to these companies not operating their own network. The closed-network arrangement allows Amex to “analyse information on Cardmember spending and build algorithms and other analytical tools that enable us to provide targeted marketing and other information services for merchants and special offers and services to Cardmembers through a variety of channels”. The exact details of how this data is leveraged are not explained, but one only has to look at the Tesco Clubcard and the advent of information driven, targeted customer marketing that it launched to see the value of customer data properly used.
The closed-network also means Amex does not have to pay an “interchange-fee” to a third-party network to enable the transaction. One could argue that incurring the expense and risk of owning a network offsets this advantage. However, the closed network model has benefited Amex in that it has allowed it to escape being subject to the the direct scrutiny of regulation across the global market targeting the interchange-fee involved in open network transactions. The interchange-fee is generally the largest component of the merchant service charge, payable by merchants but obviously ultimately passed on to the customer, and the Dodd-Frank Act gave the Federal Reserve the authority to enforce a statutory requirement that such fees be “reasonable and proportional” to the cost of a transaction. Whilst Amex has escaped the headwind of being the direct target of such regulation, it is unlikely to escape the general industry headwind that downward pressure on interchange-fees will create because Amex will likely to have to reduce its margins in order to compete.
A general concern when investing in any financial company is always leverage. Amex does have a lot of long-term debt at $59bn, which would take 9 years of operating income to pay-off. However Amex has a very good history of allocating capital and their cost of capital is just 3% on average. It also generates $6.5bn in income from its loans, against a total interest expense of $2.2bn. Another financial shock would significantly impact Amex, as it would any financial company, but its status as a bank holding company and the resulting access to low-cost Federal Reserve financing in the event of an emergency means it is well placed to weather a storm and recover.
When investing in any company The Hinterland always applies the Buffett test “would you be happy if you bought this and the stock market closed for ten years”? Generally the long-term economics of Amex are good. It is not chasing growth in the US market, it is instead looking to increase the spend per card by its relatively wealthy customer base. In the international market it is looking to increase penetration. In 2012 Amex formed 12 new partnerships to issue cards and/or acquire merchants on the Amex network, including partners in Russia and China.
The long-term ‘worry bead’ for Amex and the card industry in general is the growth of mobile and internet payment mechanisms which could eventually make significant inroads in to the traditional card payment mechanism. Amex is well aware of this and has created an ‘Enterprise Growth Group’ to pursue new forms of payments, including establishing a digital services platform, expanding alternative mobile and online payment services and forming new partnerships beyond the traditional card and travel businesses. However, the long-term concern is that people who have grown up with the internet will feel increasingly comfortable using it to transact money and this will integrate in to the internet ‘hub’ for a person’s entertainment, financial and business needs. Also, the era of ‘card snobbery’ will most likely die with the baby-boomers. Amex is better placed than most to resist this because it is highly focused on the corporate sector which is likely to be the slowest to move in to the use of internet based services such as Paypal and Google Wallet, but even the corporate sector is rapidly becoming comfortable with the security of internet technologies such as ‘cloud’ and they will ultimately gravitate towards low cost solutions if they can offer the same amount of value in terms of monitoring spend, etc, which internet based services are well capable of replicating. Therefore whilst The Hinterland would feel comfortable if the stock market closed for five years, it would be nervous if it shut for ten.
On balance, Amex is still an outstanding company and The Hinterland therefore estimates its intrinsic value (owner earnings for the next 15 years discounted to their present value, assuming no growth) at approximately $82bn. On the day of writing Amex happens to be trading at $82bn and its yield as an ‘equity bond’ is therefore just under 8%, therefore it is a hold for now.
The author is long AXP.