How To Value American Express

| About: American Express (AXP)
This article is now exclusive for PRO subscribers.

Summary

Understanding value and waiting patiently for it to occur is an important part of my investing philosophy.

There are several ways to value a company. Find the one that makes the most sense for each company.

Here are the methods I use to value a company and what my current estimated valuation is for American Express.

Point-counterpoint links to other well-conceived articles on American Express.

What is the proper value that I am willing to pay for a company?

This is more than just a rhetorical question. It should be fundamental to each equity investment decision we make. My focus is on dividends since I am already of retirement age, having begun drawing my pension back in 2002. I retired early to allow myself the privilege of being involved in the raising of our two children. The plan was that I could do some consulting on the side out of our home and make enough to pay the bills. It worked and we were able to allow our retirement nest egg to grow. Many investors do not realize that over the very long term reinvested dividends make up over 40 percent of the total return on the S&P 500. Read that again. It is important to me and may be to you (if you stop and think about it) whether you need the income or not.

What a share of stock in a company may be worth more or less to me than it is to someone else. There is no one perfect system that provides the only right value for stocks. Each company evaluates capital expenditures based upon its cost of capital and a required rate of return. Those two variables differ from company to company as each has a different cost of capital and each management has its own set of goals and perceptions regarding sustainable rates of growth. That is necessarily true because of the differences between companies and industries. A utility company is not likely to target double-digit growth in revenue because regulators that determine how much it can charge its customers would not allow it. A well-managed, mature manufacturing company may need to be happy with high single-digit growth during the good times and low single-digit growth during more difficult times. Cycles tend to define the average industry growth rates for some.

My method of valuation includes a required average total return over the long term (at least five years). Overall I am comfortable with an average return of ten percent but I vary my required rate of return by industry. For slower growing companies I tend to target eight percent but in higher growth industries I target 12 percent (and sometimes 15 percent). Generally, high growth companies also carry higher risk of loss, at least over the intermediate term, so I can only justify owning those companies if I require a higher return. But most companies fall in that middle area where I look for a ten percent return.

I find that one of the keys to investing success is to limit losses. When the value of a portfolio falls by 50 percent one must double the remaining amount just to get back to even. I prefer to buy stocks of quality companies that have a consistent history of rising dividends, revenue growth and strong cash flows. Even then, I wait for the price to hit what I consider to be a bargain. I may be stingier than you and that is fine. After all, I consider the Benjamin Graham quote to be very instructive: "In the short run, the market is a voting machine but in the long run, it is a weighing machine."

What he meant by that is that in the short run perception (and emotion) determines the value of a company, but in the long run it is the fundamentals that will always determine the true value. Fundamentals such as FCF (free cash flow) and growth in revenue and earnings, building and maintaining a moat to ward off competition for market share, returning value to shareholders and capital allocation efficiency are some of the fundamentals that determine price in the long run. If FCF turns negative or even drops to a fraction of its historical level the price will eventually follow. If revenue and earnings begin to decline year over year for more than one year, eventually the price will follow. If management grossly overpays for an acquisition (inefficient capital allocation) the price of the stock will adjust downward. The key word in each sentence above is "eventually." A stock price, or even the entire market, can defy the physical laws of gravity (or fundamental valuation) for longer than most of can imagine, but eventually everything reverts to the mean in terms of value.

Thus, when stocks become overvalued by my standards I do not chase prices in order to keep from missing out just in case the stock goes up another few percentage points. I wait patiently for the price(s) to come back to the level at which I can find bargains. Markets always correct and, as has been the case since 2000, the market can crash by significantly more than a "normal" correction of 20 percent.

Thus, I am accumulating cash and biding my time. But that is just me. Every investor needs to determine his/her own strategy.

Valuation models to consider

There are plenty of valuation models to choose from ranging from simple to highly complex. Generally speaking, the simple models have as few as one or two assumptions that the investor/analyst must decide upon (in reality, this is called guessing) to reflect future outcomes from the company and thus derive an estimate of fair value. The more complex models can require many assumptions which can also lead to either a much higher degree of accuracy or create far more opportunities for human error. As you might guess, I prefer the simpler models, but also use some moderately more complex models for validation.

As I have mentioned in the past, I like the DDM (dividend discount model) and the DCF (discounted cash flows) models. I also like to use the average historical P/E (price to earnings ratio) as a check. This would be a value based upon an earnings multiple approach. It can be argued, and rightfully so, that investors are paying now for the future earnings of a company. But the definition of "earnings" is getting harder to pin down as more than 90 percent of large capitalization companies now report non-GAAP EPS (earnings per share) instead of GAAP (generally accepted accounting principles) EPS. When I look up historical EPS for the S&P 500 by year I find tables from several different sources all with different numbers for each year. In aggregate, GAAP EPS are much lower than non-GAAP earnings and the gap between the two is rising. Therefore, I do not rely on any valuation that requires an input value for future earnings. Estimating the net present value of future free cash flow provides what I believe to be a much more reliable valuation for a company. Management and analysts can work magic with EPS but free cash flow is hard to manipulate.

I should point out that there can be significant variations in the estimated value when using a FCF approach. When revenue dips, as can often occur in a recession, FCF can fall more (as a percentage) than either revenue or operating cash flow. That is because capital expenditures needed to maintain operations may not be reduced for a temporary slowdown and it is subtracted from operating cash flow to derive FCF. Therefore, while operating cash flow may fall at a similar rate compared to revenue, FCF may fall much more. The actual long-term value of the company (its long term ability to create value) may not have really fallen so much but I like that the FCF DCF model fluctuates because it can give me signal that something is wrong earlier than some other models. When it does, it may be time to either move to the sidelines or wait for a better entry point.

A little background on American Express (NYSE:AXP)

It is important to understand the business model of credit card associations like AXP, Visa (NYSE:V), MasterCard (NYSE:MA) and Discover (NYSE:DFS). The model is different for V and MA than it is for AXP and DFS in that the first two do not issue cards while the last two do. This is an important difference. V and MA build relationships with issuing banks while AXP and DFS issue the cards directly to consumers. The issuer receives little, if any, of the processing fees charged to merchants, creating revenue from annual fees, interest, late fees, etc. V and MA keep most of the processing fee as revenue. The other (small) portion of the processing fee goes to a merchant processor (usually a bank) that contracts to enable each merchant to accept credit card payments. There may be other entities with their respective fingers in the pie but those additional fees generally are in addition to the basic transaction fee set by the associations (V, MA, AXP and DFS).

Here is an example of how it breaks down. The transaction fee is usually around 2.1 percent of the sales amount plus a flat fee of $0.10 per transaction when a merchant accepts a Visa or MasterCard credit card. For simplicity sake I am going to round it out to two percent of the sales amount to make the math easier to calculate and follow. Let's say we make a purchase of $200 at a local merchant. Two percent of that amount would be paid by the merchant ($4) and would be broken down approximately as follows:

$0.50 would go to the merchant bank that processes transactions on behalf of the merchant, $3.50 would go to V or MA, $0.00 would go to the issuing bank like Capital One or Citi.

The issuing bank may have negotiated a small cut of the transaction fee, but that is generally not the case.

In the case of AXP, it charges a higher transaction fee; more along the lines of three percent per transaction. But it also issues the card, so it collects annual fees, interest and other fees relative to the balance outstanding. Since AXP tends to focus on more credit worthy consumers who spend more it generally has a higher average transaction amount.

In the same example above where we spend $200 at a local merchant, if we use our American Express Card the local merchant bank will probably receive about the same amount ($0.50) for processing the transaction but AXP will receive almost 60 percent more ($5.50) from the same transaction.

The downside is that AXP cards are not accepted by as many merchants as V or MA cards. But AXP is making its move to become more widely accepted. That will necessitate lowering the transaction fee it charges (probably not all the way down to the V and MA level) but it will also enable AXP to issue more cards and develop more transaction volume over time. The company has a very recognizable brand globally which is a big plus. Its clientele are welcome shoppers because they generally spend more so there is a strong appeal to merchants if AXP can get the pricing right.

During the expansion effort I expect margins and revenue to be flat or even slightly lower initially. Generally, AXP derives about 50 percent of its revenue from transaction fees (which is basically all that V and MA get), about 20 percent from interest charged on outstanding balances and about eight percent from annual fees. The rest of its revenue comes from services outside of the credit card business such as traveler checks and other travel-related services, stored value products and expense management products/services. The percentages are averages that may vary from year to year. The one piece that may decline in the short- to intermediate-term is the transaction fee revenue. If it is successful, though, I would assume that the company will issue more cards (more annual fee revenue) and collect interest on higher balances. In addition, management expects the number of transactions to climb which should eventually offset the initial margin contraction from lowering the transaction fee.

AXP is not likely to issue cards to subprime consumers. The additional cards issued should come from making its card more convenient in being accepted more widely. That may make it more appealing to higher income individuals who may otherwise opt for Visa or MasterCard. That is my interpretation of management strategy and overall I believe it is a good plan.

AXP by the numbers

Looking at the chart for AXP from fastgraphs.com, we can see that AXP stock tends to begin to fall just before the recessions in 2000 and 2007. The stock hit a high of $63 in 2000 and fell to $24.20 in 2001 (61.6 percent). Then in 2007, it posted a high of $65.89 and bottomed in 2009 at a price of $9.71 (85.3 percent). In 2014, AXP made a high of $96.24 and is currently trading at $60.41 (down 37.2 percent so far). This is just for perspective and to help readers understand that it could go much further if things get really bad.

Today (Thursday, October 13, 2016 after the close) AXP offers a 2.1 percent dividend yield and sports a very enticing a TTM (trailing twelve month) P/E of 10.7. Fast Graphs calculates the normal P/E (on a forward basis) is 18.8. From that aspect the stock looks cheap. TTM EPS (earnings per share) are $5.65. So, if we multiply that number by 18.8 we would expect the price to be $106.22. However, as I look at the chart below it becomes apparent that the 18.8 P/E was a good fit prior to the Great Recession but since then it looks like a P/E of about 15 fits better. For valuation purposes later on in this article I think I'll stick with the lower P/E. Also, forward earnings are expected to be slightly lower than the TTM earnings. The lack of growth and the modification in the business model creates uncertainty and Wall Street hates uncertainty. That has led to the share price reduction. But is the stock on sale at a bargain price? We need to look deeper into the numbers.

Now let's look at some basic charts that demonstrate AXP operational results. First up is annual corporate earnings and the percent change. There is a lot of inconsistency in the growth rate in the chart below. A little occasional dip is fine but this sends up a red flag for me.

Chart by Mark Bern, CFA data source: oldschoolvalue.com

Next, we will look at annual dividends paid per share and the percent change. Well, at least this looks more (but not completely) consistent and represents the sort of chart for which dividend growth investors are looking. The financial crisis that spawned the Great Recession hit retail transactions hard and AXP shares took a beating. But the company only cut the dividend by less than eight percent. There was no growth for three years. But then the economy started growing again and people started spending, especially the well-healed, and AXP has had ample FCF with which to reward shareholders. The ten-year dividend growth rate is only 8.4 percent while the dividend has increased at an 11.9 percent CAGR over the past five years.

Chart by Mark Bern, CFA data source: oldschoolvalue.com

Now we look at the annual FCF for AXP. The good part is that it remains positive in each year. FCF dropped some in 2008 and 2009 and again in 2012, but on the whole the cash has continued to flow into the company at a healthy rate. Now let's look at what AXP has been doing with its FCF.

Chart by Mark Bern, CFA data source: oldschoolvalue.com

When FCF dropped in 2008 through 2010 the company put its share buyback program on the shelf until the economy improved to restore its balance sheet. That was a prudent move. I am not a big fan of buying back shares at peak valuations but consistent buybacks are a return of value to shareholders. That appears to be what AXP management is doing. It is sort of like dollar cost averaging, in a sense, and an approach that I approve of because as the share price drops the company just buys more shares. And that is what is happening this year.

Chart by Mark Bern, CFA data source: oldschoolvalue.com

Now I want to compare the FCF with the combined total of dividends paid and cash spent in buying back shares. Notice that in every year the company has ample FCF in excess of what it is returning to shareholders. The excess cash provides management with flexibility to invest more into the future earning capacity of the company through research and development or acquisitions.

Chart by Mark Bern, CFA data source: oldschoolvalue.com

There is one other way that management can reward shareholders in how it allocates the excess FCF. It is called paying down debt. When a company reduces its debt outstanding it essentially increases the value of the company, all other things being equal. I like what I see in the following chart.

Chart by Mark Bern, CFA data source: oldschoolvalue.com

Finally, the last chart of this series illustrates the nice trend in total diluted shares outstanding.

Chart by Mark Bern, CFA data source: oldschoolvalue.com

There are a lot of things to like about AXP, but now I want to walk you through how I value the company.

How I value American Express

Let me now look at a couple of tools that I like to use when sifting for bargains in equities. First, I would to look at the current P/E ratio relative to its historical average. As noted above I like the P/E ratio about 15 times earnings as a good fit but right now the TTM P/E is at 10.7. That P/E of 15 would yield a value of $84.75 and with the current price of $60.15 it would seem to be a bargain. But wait a minute. This is a check; not the bottom line valuation that I want to depend upon. As I mentioned above there is a lot of uncertainty around AXP in terms of both earnings expectations going forward and as to how well the adjusted business model will work. So, in this case, the P/E valuation does not seem very reliable.

Next I want to look at my DDM model using the current dividend, my expected future rate of growth and my desired rate of total return. Since the AXP share price can be very volatile during periods of economic stress I require a rate of return equal to 12 percent. My expected CAGR for the dividend is 9.9 percent. For those who are interested, I back into the growth rate by estimating what I believe the dividend will be in five years ($1.88). The result is a value of $57.07.

I also use OSV (Old School Value), primarily for historical data, and as another way to check my valuations. OSV serves up three different valuation models: DCF, Graham's Formula (best used for cyclicals, companies with volatile cash flows and growth stocks) and EBIT multiples (a form of earnings multiple valuation). In the case of American Express, the DCF model is most appropriate. OSV values AXP at $53.05 which is relatively close to my DDM valuation.

My favorite stock tool is Friedrich, partly because I spent so much time understanding its model and because it gives me a look back at ten years' worth of valuations and ratios all in one screen shot. It is also the most conservative of any stock tool that I have come across. It also uses a DCF model to value stocks and presents the output in two forms, table and chart. I will only display the table here. The table provides a quick look at ratios, valuations and trends over the last ten years. Here is the table for AXP. Notice the current value for AXP is described as "Main Street Value" and stands at $65.33 as of the last update on October 12th. That number is high relative to my DDM value and the OSV DFC model outputs. The reason is that Friedrich uses FCF and revenue growth rates rather than operating cash flow in the model. FCF was nearly flat over the TTM period relative to the prior year (2015) but revenue recovered by four percent yielding a higher Main Street Price (fair market value).

Source: AskFriedrich

The consistency in the FROIC (FCF return on invested capital) makes up for being lower than I would like to see. To make the table easier to read at a glance remember that: green is great, orange is good, yellow is average and brown is bad. There is almost always some brown on the table for most companies, but the less the better. The most important ratios that carry the most weight in the Friedrich algorithm are the FROIC ratio, CAPFLOW ratio (measures how much cash flow is used for capital expenditures), Badwill to Price ratio (helps to identify manipulation), Friedrich Cash Machine (measures FCF to revenue) and the Friedrich Equalizer (indicates whether revenues are growing or not).

Point - Counterpoint

Below are articles by other Seeking Alpha contributors. I like to offer readers points of view other than my own when considering an investment. The first three articles are more negative on AXP while the last two are more bullish. Enjoy!

"American Express Looks Weak Heading Into Earnings"

"American Express: Concerns Over Reward Costs Rising"

"American Express: Don't Pay For Amex Now"

"Use Any More Weakness In The Market To Double Down On American Express"

"American Express: A P/E Re-Rating Toward Historical Levels Suggests A 20-41% Upside"

Summary

This is not a recommendation to buy sell short shares of AXP. I believe the market, in general, to be overvalued by a significant margin which means that buying most stocks at current levels comes with far greater risk than under normal economic circumstances. I prefer to hold fewer positions in stocks when valuations are so rich. I do not believe that the potential reward warrants the level of risk I perceive to be attached to equity investing presently. There are very few companies that I can find that offer good value and appreciation potential over the next five years. I will wait to see how well management executes its new plan before considering AXP shares.

There is the argument that low interest rates and low inflation justify the higher valuations. That is true for the moment. But those conditions are temporary and will not continue forever. I am a long-term investor. That means I want to buy and hold for more than five years (generally much longer). I expect a transition to a less accommodative economic reality over the coming five years and, for that reason, am holding fewer stocks than I have in a long time. I generally do not sell unless a stock price becomes overvalued by what I estimate to be 50 percent or more. I believe that patience is one of the keys to successful investing. It all comes back to the old adage of "Buy low and sell high." I will be buying sparingly over the next year or until the bargains I like become available again. When you get to be as old as I am, you know those opportunities will come along again. It only takes patience. It is easier said than done, I know, especially for the younger investors. I have been there, too. I wish I had had more patience in my youth; I would have accumulated far more wealth if I had.

There is also the old adage that "Cash is Trash." But do not be fooled by that one. It may hold true when the market has crashed and bargains abound. There is also a saying that "Cash is King." I think this is more appropriate when valuations are relatively high. At current levels I would rather be more in cash for purposes of capital preservation. When the market bottoms again and begins to turn higher I will have plenty of dry powder for shooting those bargain fish in the proverbial barrel.

For those interested in my valuation articles on other companies, you may consider becoming a follower of mine. If you would like to receive a notice whenever I publish a new article consider clicking on the little box to the right of the follow button. If you would like to read some of my recent valuation articles you can find links to them here.

As always, I welcome comments and will try to address any concerns or questions either in the comments section or in a future article as soon as I can. The great thing about Seeking Alpha is that we can agree to disagree and, through respectful discussion, learn from each other's experience and knowledge.

For those who would like to learn more about my investment philosophy please consider reading "How I Created My Own Portfolio Over a Lifetime."

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.