The American Express Company (AXP) Valuation, Buffett Style

A must read book for any value investor is The Warren Buffett Way by Robert Hagstrom. Now in it's 3rd edition, the book profiles Buffett, his process, and deconstructs his most well known investments. It provides a valuation framework and quantifies  Buffett's famous Margin of Safety.

Unlike many of Buffett the books out there, this one is by and for investment practitioners. Robert Hagstrom is Chief Investment Strategist and Managing Director of Legg Mason Investment Counsel. The forward to the first, second, third editions are by Peter Lynch, Bill Miller, and Howard Marks, respectively. While it is not endorsed by Buffett himself , Hagstrom reports he was granted permission to quote extensively from Buffet's letters, after he had reviewed the book.

American Express is one of Buffett's best known investments, and it is well examined in the book. This is the first in a series wherein I'll review a valuation from the book, and then perform an updated valuation of the company with the same approach as the author.

AXP has under performed the market over the past year, among the reasons are the Costco deal, and the anti-trust case. 

EDIT 3/16/15: Talking with some investors brought up other risks:

• 3rd place in network size (vs V/MC)leads to competitive disadvantages over time.
• AXP could be a casualty of the disruption in the payment space, such as Applepay, ect.
• We should be asking why does anyone still want an Amex card? Is it as prestigious as it used to be?

Are these problems actually a buying opportunity?  Buffett recently opined on these issues on CNBC.

Buffett has actually gone several rounds with American Express, first with The Buffett partnership during the Salad Oil Scandal in the 1960's and then again with Berkshire in the 1990's when he infused AXP with capital via preferred shares during a cash crunch (sound familiar?).

Buffett converted his preferred to common in 1994, after AXP management sold off under performing divisions and began buying back shares. He bought more in 1995, owning almost 10% of AXP.

In The Warren Buffett Way, the Hagstrom uses a two-stage "dividend" discount model, however in lieu of dividends he substitutes "owners earnings" defined as:

	Owners Earnings Calculation
	Net Income
+	Depreciation, Depletion, and Amortization
-	Additional Working Capital
-	Capital Expenditures
=	Owners Earnings

In the 1990's American Express situation , it was approapriate to use net income, as non-cash charges roughly equal capex.

The author uses the following conservative assumptions:

• 10% owner earnings growth for 10 years, followed by 5% growth after. Management was guiding 12% to 15% net income growth at the time.
• A 10% discount rate, at the time 30 year treasuries were yielding 8%

Based on these assumptions, he concludes that Buffett determined AXP had an intrinsic value of $43.4B while its market cap was only $13B. A whopping 70% margin of safety! 

There are a host of qualitative factors to consider. The author goes into AXP's consistent operating history, and management rationality. You'll have to read the book if you want a discussion of those.

Of course, since 1994 AXP has performed very well.

To assess the current valuation we are going to carry forward the authors assumptions about owners earning's being equal to net income. Over the past few years, depreciation, depletion, and amortization has roughly equaled capital expenditures at AXP, so we are going to keep that simplifying assumption.

As for owners earning growth rate, AXP management is still targeting 12% to 15% long term growth, same as 1994. Now that is consistency!

We have been able to drive strong EPS growth within our 12% to 15% long-term target range. - Jeffrey Campbell, Chief Financial Officer- Q3 2014 earnings call, Oct. 16, 2014 

To be conservative we are going to keep our model's growth of 10% for 10 years, and a terminal growth rate of 5% after 10 years.

As for discount rates, long term treasury rates are much lower today than 1994. However, I love building conservative assumptions into my models. This way, if your valuation work points to a screaming buy, you know its not because of your optimistic assumptions. For these reasons we are going to stick with a 10% discount rate.

 One interesting take-away from the book is Buffett doesn't adjust his discount rate up or down to account for perceived risk. Rather, he demands a larger margin of safety. This is just another example of value iinvesting contradicting the approach taught in business school.

After re-running the two-stage "dividend" discount valuation model with 2014 earnings and the assumptions outlined above,  the model concludes AXP is worth $179 a share. This figure is much higher than any sell side analyst price targets. This is an intrinsic value estimate, not a price target. It also implies a 56% margin of safety at the current market price of $79. Please note, this is the book's sheet, not mine. You can get a copy of the model from the books companion website.

It's worth noting a few take away's from this intrinsic value estimate:

• The 56% margin of safety is less than the 1994 Buffett purchase margin of safety of 70%.
• The assumptions in this model are arguably more conservative than the books 1994 calculation; the risk-free-rate is much lower today, but we used the same 10% discount rate.
• To get a 70% margin of safety, AXP would have to trade at $53.
• The current price per share of $79 implies a  3% growth rate in owner earnings for the next ten years, followed by 2% growth after ten years. 
• EDIT 3/16/15: bringing the discount rate down from 10% results in some very steep discounts to intrinsic value.

Is the Costco and anti-trust news important enough compress the earnings growth rate to 2% vs management's 12% to 15% target? That is the question investors should be asking themselves.