AGB 2020.1 - Starbucks (SBUX)
AGB 2020.1 - Starbucks (SBUX)
Best in Class Store Level ROI
Subscribe to AGB - One analysis of a good business every two weeks.
Starbucks
What if we told you that you could own a store that consistently produced 30+% EBITDA margins and revenues increased 2%-5% per year due to higher volumes and prices? What if we also told you that it didn’t matter that there was an exact replica of your store being built 3 blocks away because your loyal customer base returned to your store daily to purchase your product that had almost 70% gross margins? You’d probably say that “it’s too good to be true.” Then when we showed you the data, you’d probably say “let’s build as many of these stores as possible.”
Espresso bar served premium coffee is part addictive legal stimulant, part low ticket luxury and part status signal. This socially acceptable habit was popularized in the U.S. by Starbucks after Howard Schultz (then marketing director) visited Italy in the 1980s. Ever since then, Starbucks has rapidly expanded throughout the world to over 18,000 locations in the Americas and 13,000 internationally. Starbucks consistently achieves both ticket and volume increases year after year.
What’s also amazing about Starbucks coffee is that it has the characteristics of a Veblen good. It’s one of the few consumer products that commands a higher price and still achieves higher volumes when compared to lower cost substitutes. Typically there is an inverse relationship between price and volume for consumer goods. The higher the price (premium tier) the fewer people purchase the product (exclusivity). That’s because there are fewer customers able to afford the premium tier product. But it’s also because if more people were to buy the product, it would lose the caché of being exclusive. Another product that comes close to having a similar effect is the iPhone, which in the U.S. has been able to achieve above 50% market share while still commanding a hefty premium over its Android competitors.
Starbucks stores are able to charge $5 for a specialty espresso drink or $2-$3 for drip coffee and still dominate market share at more than 60% in many markets. There are two main reasons for this. The first is that premium coffee, while materially more expensive than drinking coffee at home, is a small ticket item. That is, the masses can still afford it even if it is a daily habit for many. The second is that Starbucks coffee is also viewed as an experience and the perceived quality of that experience from its customers is very high.
Why is it a good business?
People are addicted to coffee. By the end of 2019, the global coffee market grew to $430 billion, more than half of the global cigarette market. The coffee market is growing 5%-6% annually while the cigarette market is declining because coffee isn’t bad for your health (let’s hope!) and it’s socially acceptable. Analysts are projecting the market to reach over $600 billion by 2025 as the world’s middle class achieves more wealth to afford simple luxuries. Most of that growth will come from Starbucks as it aggressively expands in many international territories.
Starbucks is a market and mindshare leader among all age groups across the world. In the U.S., 2 out of every 3 coffees are sold at a Starbucks location. Other large competitors include Dunkin Donuts and Tim Hortons as well as McDonald’s, Costa Coffee, Peet’s, etc. Even as Starbucks reaches 18,000 location in the Americas, the company is still able to grow 600-700 net new stores per year as it continues to evolve its store format and drill deeper into less saturated markets.
The Starbucks flywheel is impressive because its premium tier brand is self-reinforcing while its digital strategy complements the brand nicely. More customers buy Starbucks coffee, which increases the value of the brand, which then leads to more frequent purchases. This coupled with a consistent consumer experience leads to attracting more customers. Starbucks was able to complement the brand with its digital strategy. Starbuck’s rewards program was introduced in 2008 and its mobile app in 2009. These two offerings lower search costs and increases value for its most frequent customers. Overtime this leads to more frequent purchases and in turn increases the value of the brand.
And of course as the company gets larger, there are scale economies achieved. Starbucks can better spread its fixed costs over a larger base of stores, which leads to more customers.
Starbuck’s crown jewel might be its mobile app and rewards program. The mobile application helps customers track their rewards status and order coffee in advance of going to the store. Mobile orders increase utilization rates of stores, leading to higher profit margins. Mobile orders made up 5% of orders in 2016 and grew to 12% of orders by 2018. In 2nd quarter of 2020, mobile orders grew to 16% of total orders.
The rewards program has also grown nicely reaching 15.3 million members by 2018. That number is 19.4 million by the 2nd quarter of 2020. The rewards program is matched with its prepaid card program. The prepaid card program has allowed the company to access interest free capital over many years. In FY19, the rewards program made up $1.1 billion of its deferred revenue. This means in the past fiscal year, customers have floated the company 72% of its inventory costs. This allows Starbucks to operate with a lighter working capital load, which facilitates the ability to quickly buildout new stores.
Returns on capital?
The restaurant level returns for Starbucks leads the industry. In the U.S., a store costs under $700,000 to build and will produce roughly $1.5 million in revenues with 30% EBITDA margins once it’s up and running. High margins leads to a low payback period of 1.5 years. A licensed store owner would have to pay a 15% royalty on top of that, which implies the payback period is roughly 3 years. That’s still great returns with relatively low risk, given that it is the category killer and coffee addiction isn’t going anywhere soon. The returns are even better in China and this may be part of the reason why Starbucks decided to buy the East China licensed stores back in 2018.
Starbucks has the ability to raise prices should it be required. In keeping up with inflation, prices have gone up 2-5% annually in the Americas and 1-2% internationally over the past 10 years. Price fluctuations related to inputs (coffee beans, transport, utilities, etc.) can be passed down to the customer, evidence of pricing power.
If you look at the long-term view of the enterprise, Starbucks has been able to generate a great return on its incremental capital invested in the business. From a unit economics perspective, we know that the payback period is 1.5 years on a store that is fully up and running so let’s estimate 2 years to reach breakeven. That would imply a 50% return annually, at the store level. Of course Starbucks has other expenses to run the enterprise such as general and administrative costs, interest expenses and taxes. There’s also working capital requirements to grow the business, which the company has been able to manage well with its rewards and card program.
At the company level, incremental returns on capital are near 35% over the past decade, though there is volatility around these returns as the company optimizes the company owned versus licensed store mix within each territory. Starbuck has repurchased licensed stores in certain regions or sold off certain company owned stores in others.
Reinvestment potential?
A great exercise when determining the reinvestment rate of a business is to see what the ceiling looks like and the time frame it would take the company to reach it. This is a blue sky scenario, but it’s good to know a company’s addressable market before estimating a long-term growth rate. At its current state, Starbucks is almost near saturation in the Americas. Net new store openings will range between 600-700 per year or a 3%-4% growth rate going forward. Locations within the U.S. that are less saturated are mostly in the Midwest and Southern regions. Internationally, Europe still has some room to grow, but should see deceleration from double digit unit growth in recent years.
China will be the largest driver of growth internationally. The company has an internal target of reaching 6,000 stores by 2022 from its current base of 4,300 stores. The long-term potential in China could be bigger than that of the U.S. given the population disparity, growth in overall coffee consumption as a culture and the rising middle class doubling by 2022 to 600M people. Coffee consumption in China is so sparse (even with the many Starbucks locations) that the cup per capita consumed is less than 1 per year, whereas it’s over 300 in the U.S. We anticipate that it will take years to change China from tea to coffee as the preferred medium for caffeine intake, but growth should be meaningful nonetheless. With that in mind, China’s store count should increase at an average rate above 33% for the near future and then taper off, leading to low double digit growth for all of Asia Pacific.
We estimate that international should grow between 9%-11% annually and reach a similar store count to that of the Americas by sometime near 2024. Even at this growth rate, international Starbucks locations will only match that of the Americas by 2024. We estimate that international regions have the potential to be more than 60% of all Starbucks locations.
Over the medium-term we expect that enterprise level revenue growth to be 7%-11% annually, with 5%-6% coming from store growth and 2-5% from comparable sales growth. This level of growth requires a reinvestment rate of approximately 35% of the company’s operating cash flow each year, which we estimate will achieve returns near 35%. Using those numbers as a baseline, we estimate the current trajectory of Starbuck’s medium-term growth in intrinsic value is 13% per year.
What else is important?
License vs. store owned
One thing to consider when analyzing food service companies is the idea of franchising. While Starbucks doesn’t have a franchising program, it does allow others to license the brand. Licensees are responsible to pay for start-up costs as well as the material inputs to run the store (COGS) and the ongoing costs to run the store (opex). Starbucks provides the coffee and food related products for resale and equipment for licensees to use for their operations in exchange for royalty and license fees at roughly 15% of revenues. Licensees must follow the protocols set by Starbucks and many employees are directly trained by the company.
Whether Starbucks chooses to license versus own the stores outright depends on the market and available locations to the company. In the Americas region, 45% of the stores are licensed and in international regions, 55% of the stores are licensed. Many times a licensee may have access to better retail locations than Starbucks could acquire, especially in certain international regions.
So what if Starbucks chose to go solely with the license model? Instead of having to operate 16,000 stores, Starbucks could sit back and let the licensees do the heavy lifting. From a financial perspective, this would shrink the company’s revenues by almost two thirds but increase operating income by almost 80%. In theory this would result in a massive increase in equity value for the company.
However, without directly operating company stores, Starbucks would lose leverage on the purchasing power of raw materials like coffee, tea, ingredients for food, etc. Remember that Starbucks controls all the purchasing, roasting and packaging of their coffee. They could still do this with an all license model, but it would eat into margin. Furthermore, it would be difficult to continuously improve store operating procedures and human resources related protocols without having a direct touch to how stores operate. Innovation in store formats would remain stagnant, something the company has continuously improved.
Saturation of the market
A big question to ask is “when is the market saturated?” We think most people would agree that having three stores in one city block could be too much. New York City has the highest number of Starbucks locations at 241 in 2019. Excluding the 10 from Staten Island, 231 stores are in the other four boroughs, Manhattan, Brooklyn, Bronx and Queens. Those four boroughs are 244 sq miles, meaning New York (ex. Staten Island) has almost 1 Starbucks store for every square mile.
Another way of approaching this question is population density. According to Mercury News, Las Vegas has one Starbucks for every 5,133 residents. Understanding that there are a lot of tourists that travel to Las Vegas, we look to 2nd on the list which is Burbank, CA with one Starbucks for every 5,242 residents.
Using these two at a guide, let’s look at the opportunity in China. Shanghai is 6,218 sq. miles and one of the densest cities in the world at 24.3 million residents. If we use New York as a guide in land size, Shanghai could theoretically hold 5,886 Starbucks locations. If we use Las Vegas or Burbank as a guide in population size, Shanghai could have capacity for 4,673 locations. It’s estimated that Shanghai has just over 250 stores or less than 5% of its theoretical capacity. The largest cities in China are all under penetrated if we assume that coffee consumption rises like it has in the U.S.
It’s not just the big cities that have growth potential in China. Pat Grismer, Starbuck’s CFO mentioned on its Fiscal 2019, 4th Quarter earnings conference call that,
“our store openings in lower tier cities in China accounted for a meaningfully higher percentage of total store growth in that market versus the prior year. Yet portfolio investment returns remained very robust, demonstrating Starbucks resonance with China's growing middle class.”
Where is optionality?
It’s difficult to imagine Starbucks discovering a second S-curve product like many of the big tech companies have. Microsoft’s first success was Windows, but had subsequent successes in Office and Azure. Google first success was Search, but had successes in Android and YouTube. Facebook’s first success was Facebook, but had success in Instagram and WhatsApp. Amazon’s first success was online retail, but had success in AWS.
Optionality is a hard one for Starbucks because the company has tried to branch out into tea in the past and failed. There is excess capital being generated at the company and it’s being returned to shareholders in the form of dividends and buybacks. Large acquisitions wouldn’t make much sense with the core Starbucks brand. Where we do see limited optionality is in creative partnerships with other food or coffee companies such as the Global Coffee Alliance with Nestlé.
Where Starbucks could have had a second S-curve product was in payments. Starbucks launched its prepaid card program in 2008 and combined it with its rewards program in 2009. Users of the Starbucks rewards cards essentially use it as a prepaid debit card that’s only accepted at Starbucks locations. For using the card, members get certain rewards to apply to future purchases.
The real opportunity was for Starbucks to leverage the member data and the members’ cash infused in the system to create a prepaid card network that was accepted at all merchants. Instead of making it a Starbuck’s only closed network, they could have partnered with Visa or MasterCard to allow members to use their Starbucks cards everywhere Visa or MasterCard was accepted. Starbucks had a substantial lead versus the big players today. Cash App was started by Square in 2013 and Venmo was started in 2009 and was acquired by Braintree (PayPal) in 2012.
If you made it this far, I hope you received some value from reading our analysis. Please subscribe to the free newsletter (if you haven’t already) and share with anyone that would find it valuable. Thank you for your support!