I think most investors agree that Starbucks (SBUX) is a fantastic company, but as an investment it is not quite as popular due its P/E-ratio of 28.4. However, I believe that the P/E-ratio is highly misleading, as it ignores risk and the future growth rate.
One way investors can take those factors into account is trough a discounted cash flow (DCF) analysis of the company. The disadvantage of the DCF-analysis is that it's very time-consuming and one needs to make a lot of assumptions. But I am of the conviction that there are assumptions behind every investment decision, and the difference between a great and a mediocre investment lies in the quality of those assumptions. So in this article I will present my reasoning behind my estimations, and I will argue why Starbucks (despite the high P/E-ratio) is a great investment for the long-term investor.
Future earnings estimations
In previous articles I have analyzed every segment which Starbucks operates in, and I have estimated the operating profit for the next 10 years. To those who have not read my articles, you can read a short summary below:
- CEO Howard Schultz turned Starbucks around from 2007-2009 through a huge restructuring. He improved the supply chain as well as the brand value of Starbucks.
- The Americas segment is Starbucks' largest segment and it is extremely profitable. But there is still room for new stores, and Starbucks is in the process of extending its product line with new offerings such as bakeries, juices and energy drinks.
- Starbucks has some problems in the Europe/Middle East/Africa segment, though those problems aren't particularly related to the financial crisis. Rather, Starbucks hasn't adapted to the cultural differences in the countries, but going forward, I believe the company has learned a necessary lesson, and I expect an improved operating margin in the segment.
- Starbucks has had success in Japan and they are currently in a process of adding a lot of new stores in China, which has driven growth over the last quarters. Going forward, Starbucks plans to double the current amount of stores in the country within 2015, and it will be expanding in India as well.
- The Global Consumer Product Group consists of products which Starbucks sells through retailers. This group is the highest growing segment of Starbucks, and it has a bright future, as Starbucks recently released a new coffee machine, the Verismo.
Below, I have graphed my future earnings estimates of the various segments.
Estimating working capital and CAPEX
To get from earnings to free cash flow, we have to subtract the change in working capital and CAPEX from the after-tax operating profit.
Working capital is calculated as: Current assets - Current liabilities. As investors we generally prefer companies which minimize current assets and maximize current liabilities, as it implies that the company has a business model in which it receives money before it pays its suppliers or other short-term creditors.
In the below graph we can see that the current assets of Starbucks have increased a lot over the last couple of years, which has had a negative impact on the free cash flow.
If we look at CAPEX (which, unlike depreciations, impacts the free cash flow), we can see that it has decreased (as a percentage of revenue) since 2007. Normally, one would expect growth companies to have higher CAPEX than depreciations. So the trend in the last couple of years has (unlike the increase in working capital) benefited the free cash flow.
Looking forward, I expect CAPEX to increase as a percentage of revenue, while current assets as a percentage of revenue will decline gradually over the coming years. Management expects CAPEX of 12% of revenues in 2013. Afterwards I expect this percentage to average roughly 9%, as Starbucks plans to open a lot of new stores which will increase CAPEX.
The discount rate
As earnings today are more valuable than future earnings, investors need to discount the future free cash flow to obtain the present value. But the question which then arises, is; How do we determine the discount rate?
I think there are three different approaches one can use.
Approach 1) Use the CAPM method (discount rate = risk free rate + risk premium * beta) and estimate the beta by looking at historical volatility. Yahoo Finance reports a (historical) beta of 0.88.
Approach 2) If historical volatility isn't a good indicator of the future risk of the company, one could make a qualitative risk-analysis of the company and use it to estimate a "fair beta."
Approach 3) Acknowledge that neither approach is particularly accurate at estimating the "true beta," but one could combine both approaches by averaging approach 1 and 2.
I believe approach 3 is the best approach, and therefore I will use it in my DCF-analysis.
I think the most important answer one needs to answer in a qualitative risk-analysis is how confident we are in our earnings estimations. Some companies (e.g. McDonald's, Procter & Gamble, Coca-Cola) have a reliable business model and we can with a relatively high degree of certainty estimate their future earnings. Most tech companies, however, have very volatile earnings, and hence they should be given a high "fair beta."
Regarding Starbucks, I think the Americas segment is very comparable to McDonald's and Coca Cola. Starbucks' customers aren't going anywhere over the next 10-15 years, and I am confident that Starbucks will grow revenue between 5-10% on an annual basis. EMEA is much more uncertain, but as the segment is so small, it barely matters. Regarding China Asia Pacific, the exact growth rate is difficult to estimate. It could just as well grow at 10% as 30% at an annual basis over the next 10 years. But regardless, it's going to grow at a pretty good rate, and Starbucks has already succeeded in obtaining a strong brand in the segment. The Global Consumer Product segment is somewhat comparable to CAP. The exact growth rate is difficult to estimate, and there are operational risks related to the use of 3rd party retailers.
I'd conclude my short qualitative risk-analysis by listing (what I believe to be) fair betas for each segment:
The Americas segment: Fair beta of 0.55.
EMEA: Fair beta of 1.1.
China Asia Pacific: Fair beta of 1.
Global Consumer Product: Fair beta of 1.15.
The fair beta of Starbucks (using appropriate weights) is roughly equal to 0.84. This means that the average beta of Starbucks (combining approach 1 and 2) is ((0.88+0.84)/2) 0.86.
In the below tables, you can see my future expectations for free cash flow and the fair value estimate.
As you can see, I believe that the fair value of Starbucks is $59.56. In terms of alpha, Starbucks will be a decent investment, though not a huge generator. But I still like this company a lot, as it provides a superior alternative to long-term government bonds.
Given my free cash flow estimations, the implicit yield is 8%. So why would you be satisfied with a 2% yield from 10-year or 30-year government bonds when you can get a return which is 4 times higher by investing in Starbucks?