Friday, April 10, 2009

American Express - stock analysis and review

Recent share holder letter provides some valuable insight. Here are some that will interest long term investors.

1. Consumer spending pattern: The most important driver of value for AXP is consumer card spending. The US consumer card spending reduced 12% in fourth quarter of 2008. International card spending showed some growth but much less than last year. The reduced spending was particularly visible in luxury goods which is especially important for American Express. The key question is how much of this decline is attributed to secular shift in the consumer spending patterns. Are consumers temporarily withdrawing or will we see a cautious consumer whose prefers to save for next few years. By all accounts, it seems that we will have a cautious consumer at least for next few years. Since American Express depends on consumers spending and transactions, this can have a significant impact on revenue and bottom line.

2. Underwriting decisions: Management is candid about some of their actions that resulted in increased credit loss. First, AMEX card member base is skewed towards states hit hard by real estate crash. [Btw, I understand Discover was more cautious in this regard.]. Second Amex added more card members than industry (higher growth). Third, they have more small business accounts. These decisions are elaborated below as they are significant in my view.

3. Floor for credit losses: The above three actions illustrate that Amex fully participated in the credit boom. This resulted in significant growth in good years which is coming back to bite in bad years. The net result of above actions resulted in middle of the pack past-due and write-off rates (bad loans that cannot be collected). Further, this loss cycle is far from over since unemployment is expected to continue over many months to come. The exact level of credit card losses is difficult to predict in this environment. Historical indicators may be misleading as the current credit bubble induced recession differs from either the valuation driven tech bubble or a general economic slowdown without structural issues. It is difficult to estimate future losses and considering credit card loans are unsecured loans, uncertainty definitely exists in this regard. It is for this reason that I believe that valuing Amex at this point is not possible with any level of certainity.

4. Business model: American Express traditionally commands a higher multiple in the market for a reason. Historically, American express business model is based on spending by high networth customers who will pay the bill rain or shine. The focus is on discount fees from merchants and membership fees from card holders. Increasing balance was not the primary focus. This meant that Amex had low credit risk and market rewarded this handsomely with high multiple. [There is a reason Banks have low multiple for same earnings. One is leverage and other is possibility of credit loss.)

The question to be evaluated now is whether the Amex business model has changed as a result of expansion past few years coupled with poor underwriting decisions (outlined in #2)? The write-off rates are in the middle of the pack for a top quality company. If American Express continues to expand in good times and creates higher write off rates in poor economy, it will be treated more like cyclical business and will not justify premium multiples (correctly so). This affects long term share holder value. The key issue is whether American express moves back to a business model of low credit risk. I suspect Warren Buffet did not anticipate holding a company with middle of the pack write-off rates. If so, that will inevitably mean more subdued growth in the future.

5. Risk conscious culture:
The net charge off being in the middle of the pack for a top class company like American express is unacceptable to say the least. Was there too much focus on growth? How did the company miss out on risks during real estate bubble when at least one competitor recognized it? For a financial company, growth can be dangerous if it is not accompanied with sound underwriting. I understand American express has tweaked their models in response to the crisis but it is equally troubling that the culture within the company did not catch this earlier. Holding a financial company is meaningful only if it has risk conscious culture. Any deviations can be fatal and as well difficult to rebuild the culture. I do not see anything else that has changed other than tweaking models. Note models are a result of human decisions anticipating risks not the other way around. This point is key.

6. Compensation: I tried to read through the proxy for compensation details. It was complex to say the least and based on many parameters. I think it should be simple enough and depend on few critical drivers of the business. Some combination of underwriting profit growth and write-offs over a cycle would make sense. In a year like 2008, anything other than base pay is excessive.

7. Credit card securatization: It is an open question as to how the credit card receivable securatization market will function after the current crisis is past? Could it be relied upon for funding? Since credit cards are unsecured loans, it is always possible that this market will freeze whenever investors are nervous. This risk applies to all credit card companies. In this regard, American express has brand advantages to raise capital from other funding sources. However, as a whole, the total credit issued will be reduced in the future as securatization market will demand higher credit quality.

8. Cost expenses: American Express has announced 10% reduction in work force. I think this is in keeping with realistic view of future growth after years of expansion. However, it also shows that future growth is going to be subdued.

9. Regulation: New regulation is inevitable and it is not going to be friendly to credit card companies. It is difficult to predict the nature of this. I suspect regulation will affect the credit card companies whose business model is based on balances more than AMEX.

Conclusion: I think future growth of Amex will depend on a number of factors outlined above. At this time, AMEX is in "too hard to value" pile. Here is a recap why it is so.

First, Amex has to survive the credit losses in the short term without raising expensive equity capital. I do not think we have reached peak write offs and that anyone can predict this with any accuracy since economy is still in poor shape.

Second, will AMEX learn from this downturn and go back to the business model that results in low credit risk across entire cycle? It is an unknown. This will decide the multiple market provides (correctly so).

Third, credit card industry has a number of external factors that can influence it's future like regulation, consumer deleveraging, securatization market etc. These are hard to predict.

Fourth and most important, will management create an environment that is focussed on sound risk management? This is the most important aspect of a financial company and in last bubble risk controls failed the company. Without this, owning financial company is a gamble.

Considering all of the above, I think it is not possible to meaningfully value the company at this point. As a disclaimer: I do own Amex shares and plan to continue to hold. Please leave your comments by clicking the comment button below.

4 comments:

Anonymous said...

At what price would you consider a buy?

Sriram said...

I have had many questions from friends about what price is the stock a buy?

It is a good question as it determines when you buy. I will take a shot at it realizing that it is "difficult to value" pile and hence associated uncertainity.

1. First, if Amex makes significant changes to business model to take less credit risk, it will get higher price multiple. I will assume for now this is not the case as it is difficult for most managements to pass up earnings growth. That being the case, we will assign a P/E like a bank (say 10 to 14). Note that it is not a bad multiple for middle of the pack write off's. [Before the crisis market did not view AXP as equivalent of a bank and allowed higher multiple.]

2. The next question is to settle on a earnings figure once crisis is past. This is much more difficult to predict. Lets say consumer deleveraging is real for next few years. Each card that defaults is one less card that will contribute to earnings in the future. May be it results in healthy earnings! Plus, likely AXP management will get somewhat conservative and grow less. Plus credit card securatization market may not allow anything less than good credit quality for next few years. Plus real estate bubble which contributed to growth in bubble is not coming back anytime soon. Keep these in mind as we estimate the earnings next!

3. In 2007 (roughly bubble period), AXP earned $3.36 per share. In 2011, lets say the earnings drop by 30% from bubble years because of above factors. So it comes to $2.25 per share. Slap a multiple of 14, it comes to $32. You can try other multiples.

4. Mr. Buffet said, it is a hell of a buy at $10 and that would still be the case if above outcome is realized. Since the bubble in credit card industry is not yet completely over, caution is prudent.

Hope this helps. Please leave your thoughts.

Stock Analysis said...

Thanks for Sharing this descriptive information about Stock Analysis. This type of communications are really very helpful in upgrading and improving your trading & investment skill. For more such type of information, Kindly Visit: http://www.smcindiaonline.com/stocks-research.aspx

Anonymous said...

Hello there, just became alert to your blog through Google, and
found that it is truly informative. I am gonna watch out for brussels.
I will appreciate if you continue this in future. Numerous people will be benefited
from your writing. Cheers!

Feel free to surf to my web blog agoda coupon codes