A few weeks after I came back from a trip to Italy in August, 2018, Starbucks (SBUX) opened their first Italian branch in Milan to a crowd of curious, but critical longtime espresso drinking Italians. For many of us, whether you drink their beverages or not, Starbucks needs no introduction, and the coffee giant is here to stay.
But is it a value stock?
One type of asset value investors look at are companies like SBUX that have shown an ability to persevere through the vicissitudes of the market because of slow and steady expansion of their moat and competent, shareholder-friendly management. Buffett himself never shied away from investing in giants like IBM (IBM) and Johnson & Johnson (JNJ) when the price was right.
The entry price of a company’s shares is one of the keys in value investing, but don’t take it from me:
“Long ago, Ben Graham taught me that ‘Price is what you pay; value is what you get.’ Whether we’re talking about socks or stocks, I like buying quality merchandise when it is marked down.”
– Warren Buffett
So what is a fair price for SBUX, and is today’s price a good entry point for the company? In this article, we’ll use a couple of Old School Value’s (OSV’s) analytical tools to determine just that starting with a classic DCF, and then moving on to an EBIT multiples valuation. Finally, we’ll be able to come to a fair value for SBUX.
Grab a cup of coffee (or tea), pull up a chair, and let’s dive into the analysis.
Discounted Cash Flow (DCF)
Opening up SBUX in OSV’s DCF part of the app, we are greeted with the following default output:
According to the defaults, the fair value of SBUX is $203.35 with a Margin of Safety of 65% on the current price of $70.84 (at the time of writing). If this is true, that would make SBUX a screaming buy. But we aren’t passive investors, and we want to break into the data. Let’s alter the inputs a bit and see how it changes.
First, let’s take a look at the DCF style and starting value. Since DCF is attempting to predict cash flows for a long time into the future, we are generally going to want to choose the metric which shows the most steady, predictable growth. We have three choices: FCF, Owner Earnings FCF, or EPS.
Looking at the cash flow data, we can see that FCF and Owner Earnings, while for the most part growing over time, are quite erratic and even negative in some years (more on that later). Remember, we want to see clean growth over time if the model is going to be accurate in its predictions. EPS looks to be the best of the bunch at $3B, but using earnings to predict cash flows can be dangerous since earnings can be misleading, or worse, manipulated.
If we want to avoid using EPS for our projection, then we’ll need to do a bit more digging into these numbers to find out why they are so erratic — why are FCF and Owner Earnings up at $10B TTM while EPS is down at $3B?
Opening up the Income Statement in OSV, we can see a $1.8B non-operating gain for 2018. A huge jump from previous years, which probably had an impact on earnings.
SEC filings can be scary things to the uninitiated and quite daunting to work through unless you’re well-versed in financial terminology. But a little practice makes perfect, and it’s the only way we’ll be able to find out the story behind the numbers. To further investigate, we’ll need to take a look at SBUX most recent 10-K filing to get the details on the non-operational gain.
It turns out they had a 50/50 joint venture (called “East China” in the 10-K) based in Hong Kong that they bought out for $1.4B back in December, 2017. They reassessed the value of their original 50% holding at the same $1.4B. This figure, along with a $400M sale of Tazo, adds up to the $1.8B gain. Aha! But the $1.4B is a non-cash gain, and the $400M is a one time gain. Will these affect our DCF?
Yes and no. The sale of Tazo means that those cash flows will be gone ($56M according to the filing). The acquisition of the East China group should translate into future cash flows for SBUX to the tune of $73M/yr in years to come. But for a massive company bringing in $3B in cash flows, these figures are rounding errors at best and can be largely ignored for our DCF.
But this still leaves us with the mystery of the erratic FCF and Owner Earnings. Where is the $10B coming from when EPS is sitting at $3B? Let’s go to the Cash Flow Statement next and look for big changes:
We can see a huge $7B increase in “Change in Deferred Revenue.” The answer to where this is coming from is the first sentence in the link to the 10-K provided above:
We entered into an agreement on May 6, 2018 to establish the Global Coffee Alliance with Nestlé. On August 26, 2018, Nestlé licensed the rights to market, sell and distribute Starbucks consumer packaged goods and foodservice products in authorized channels. We received an upfront payment of approximately $7 billion primarily of prepaid royalties which was recorded as a liability to current and long-term deferred revenue and will be recognized as other revenue on a straight-line basis over the estimated economic life of the arrangement.
Although this means the company received a lot of cash from
Nestlé, this is not going to be a recurring event, so it needs to be deducted from our cash flow projection. In the end, it looks like we can use the FCF number minus the $7.1B (the $1.8B from before is already subtracted out of the cash flow).
This leaves us with a number around $3B, which is exactly what EPS says in the first place! However, it doesn’t always work out that way, and it can save us a lot of headaches and regrettable investments later if we check.
After changing the DCF starting value, we have to consider what the growth and discount rates should be. When thinking of growth rates, we need to confirm whether the rate given by the app (12.54%) is realistic or not.
Although analysts currently expect EPS to grow 12.54%, this seems high to apply to cash flow. Revenue and EPS have been increasing, but not at 12.5%, and EBITDA has been flat for the past two years. The EBIT margin has declined slightly to around 14%, too, so it’s hard to see where double-digit growth is going to come from.
Starbucks is not the young and budding company that it once was. While there are growth opportunities still to be had, and the CEO Kevin Johnson says they are well positioned for further growth in China, this is all speculation at this point.
Based on the numbers for growth we have, I don’t think it makes sense to apply a 12% growth rate over the long term. So what would be a reasonable assumption of future SBUX growth? Let’s look at three possible scenarios, and see what that does to our margin of safety.
At 12.54% default Growth: 13% MOS
At 10% Growth: 0% MOS
At 5% Growth: 0% MOS
So if the growth slips even 2% from estimates, then we’ll be left with no MOS at all. In other words, there is no room for error going forward with 12%.
For now, we’ll go with an 8% growth rate in our final analysis — somewhere between the optimism of 12%, and the pessimism of 5%.
Discount rates are a controversial topic. Some say to start with the risk-free rate and adjust from there. Some say that you have to choose a discount rate dependent on the industry. And yet others say that it should be calculated based on a calculation of WACC (Weighted Average Cost of Capital). Since there is no perfect, universally accepted way, all we can do is make an educated guess. I like to start with 10% as my minimum required rate of return, and then adjust upwards for companies that are young, volatile and unpredictable (TSLA, SNAP) or downwards for well-established companies with big moats (MCD, AAPL). So for SBUX, I would say that 8% is fair.
My short answer to why I don’t spend a lot of time calculating the discount rate:
“It’s better to be roughly right than precisely wrong.”John Maynard Keynes
Plugging all of our assumptions in, we get the following:
After our small changes to make the valuation a bit more conservative, the fair value has dropped precipitously to $59.07. This is likely more in line with reality, but we now have a 0% margin of safety (MOS). I normally like to see this at 25%, even for a huge company like SBUX.
As a side note, whenever I use OSV to do this sort of analysis, I try and keep the fair value numbers out of my sight until I’m ready to input everything. That way, I can mitigate any bias I might have about a particular company and avoid changing numbers to make the model say what I want it to say (never a good thing!)
For now, we will put the numbers for the DCF analysis aside and come back to them again later after looking at another analytical tool.
Earnings Before Interest and Taxes (EBIT)
Now, we’ll see what the numbers in EBIT say about SBUX and make a few adjustments there as well. First, the opening page for EBIT with all of the default OSV-generated values, which is painting a much less rosy picture of SBUX fundamentals than the default DCF was:
For EBIT, we are going to need a few numbers: the Revenue estimate, and three Operating Margins and EBIT multiples, one for each of the different scenarios above.
We’ll leave revenue as it is, but we are going to fiddle with the other two numbers. Let’s take a look at the income statement to check the operating margin, shall we?
From this information, we can better construct operating margins for our three scenarios. Back in 2013, SBUX was in the red due to a legal dispute with Kraft (KHC) which cost them 2.8 billion dollars. Along with acquisition of Teavana and other investiture, this drove them into the negative. Compared with other years, however, this is not a “business as usual” number for the company and can be safely ignored. However, the margins have been decreasing a bit for the past couple of years.
My goal here is to grab two numbers that would reflect both a “conservative” and an “aggressive” scenario. I’m comfortable leaving the “normal” scenario to the app (15.61%).
For the conservative scenario, let’s assume SBUX continues to lose ground with their operating margin and it drops down to 13% – below even the TTM of 13.9%. Meanwhile, an aggressive investor might take the high end of the past few years, 18.1%.
Moving on the the EBIT multiples, SBUX tends to trade at an EBIT of around 21x. Let’s look at the data.
This time, I’m going to change all three multiples. The minimum that SBUX has been in the past 5 years was 20.56x, back in 2014, and the highest was the year after that in 2015 at 26.48x. To smooth these numbers out, we can look at the median of 21.31x and use that as our normal case, but why not make it a nice round 21x?
For the conservative case, let’s go further than the 5YR minimum and assume a 10% lower than average scenario of 18x. On the other end, we’ll use the 5 year max of 26.48. This is a huge increase, but judging by the TTM of 26.41, it is not impossible. Again, I like to use round numbers for consistency, so let’s drop it to 26x.
After typing all of this into the model, this is what it comes up with:
So what’s the takeaway?
Let’s quickly review the results of fair value from each of the models.
DCF: $57.03 (MOS: 0%)
EBIT: $63.77 (MOS: 0%)
When using multiple analyses, I like to average the values together at the end to get a more nuanced fair value. Doing this gives us:
Fair Value (Avg. of above): $60.04
At the time of writing, the price of SBUX is sitting around $70 or overvalued based on this analysis.
What do you think about SBUX? Would you invest at today’s price?
Disclosure: I am long SBUX.
Jake Reed is a DIY investor who has been using OSV and its resources to analyze companies since 2010. He lives and works in Japan.