1. Investment Thesis
We believe Under Armour (UA) common shares are overvalued; possessing unrealistic growth prospects, poor earnings quality, and no real competitive advantage. To date, this thesis has lost, but has not been falsified. We believe the institutions, which hold 87% of the float are impressed with UA's growth and have bid up shares.
We desired to perform an analysis of UA with no particular cognitive bias. Admittedly, we failed our goal of maintaining an even viewpoint. To mitigate this failure, we have taken objective approaches where applicable and been overly aggressive in our assumption-making (given our negative bias). As value-oriented investors, we adhere to the margin of safety principle explained by Graham. Margin of safety should mitigate against things such as, but not limited to, model risk, cognitive bias, and the vagaries of time.
To value UA on a per share basis, we used an expected value (EV) approach. We simulated multiple scenarios, weighted their values based on a priori probabilities, and summed them to get an EV per share. Three scenarios modeled were US, global, and best case. We also used an analyst scenario for benchmarking purposes which uses assumptions that research analysts and the market are currently pricing into UA's share value. The primary difference between scenarios are the revenue assumptions, which will be detailed later. EV per share came out to $25.69; providing a margin of safety at 56%.
With such a large dispersion from intrinsic value, we suggest going short UA. In regards to what investment tool to utilize, we suggest the underlying. We have not found a catalyst to begin a negative feedback loop in the near future,making wasting assets sub-optimal. The indifference to recent insider selling further exasperates the lack of desire to use leverage.
2. Business Summary
Under Armour was incorporated in Maryland in 1996 and filed for IPO in 2006. According to the S1, sales grew from $5 million in 2000 to $205 million in 2004. The growth has remained strong since going public. From 2006 to 2011 net sales have increased from $430 million to $1.47 billion.
Under Armour's principal business activities are developing, marketing, and distributing branded performance apparel, footwear, and accessories. Its products are sold to men, women, and children. The moisture-wicking fabric apparel is the primary driver of UA's sales. The companies target market is primarily athletes and people with active lifestyles.
Apparel sales represented 76% of total sales. This has come down from 2006 when apparel represented 90% of net sales. The apparel is "engineered to replace traditional non-performance fabrics in the world of athletics and fitness with performance alternatives designed and merchandised along gearlines." The three main gearlines are HEATGEAR, COLDGEAR, and ALLSEASONGEAR.
The company began offering footwear in 2006. Footwear sales represent 12% of total sales in 2011. The trend in footwear is not forthcoming, as it grew substantially into 2009 but suffered in 2010. Sales are primarily in cleats, but also in categories such as running and basketball shoes.
Accessories, which include hats, bags, and athletic gloves, represent 9% of sales in 2011. This sales category has historically been a low growth driver for UA. In 2011 there was rapid increase in accessory sales due to the sale of hats and bags.
In a distribution sense, the majority of sales (70%) are from wholesale channels such as national and regional sporting good chains and department chains. Dicks Sporting Goods and The Sports Authority accounted for 26% of total net sales in 2011. Direct-to-consumer represented 27% of sales.
Geographically, North America represents 94% of sales. The company has expressed the desire to expand in other areas. Its ability to achieve this goal has not been shown. This issue will be discussed in the corporate strategy section.
3. Industry and Competitive Analysis
Apparel and Sports Apparel
U.S. Aggregate Apparel was the first industry data analyzed. According to statistics from the 2008 American Apparel & Footwear Association, U.S. consumption for apparel has decreased substantially over the past decade. "The market for apparel and footwear continues to underperform compared to the economy as a whole" (page 3). The average growth rate from 1997 to 2008 was 2.1%. Data conducted by BEA.gov shows that clothing and footwear had negative growth in 2008 and 2009, but bounced back in 2010 and 2011. We were unable to draw significant conclusions, except for the belief that performance apparel will diverge to aggregate apparel rates in its perpetuity.
The global performance apparel industry is expected to grow from $122 billion to $126 billion from now until 2015. Awareness in having an active and healthy lifestyle has been increasing in recent years. Sales have followed this trend. We expect the performance apparel market to grow at a faster pace than overall apparel market.
There are attributes of monopolistic competition and oligopoly in the performance apparel industry. More recently, the industry has taken on more monopolistic competition attributes such as:
1. "Number of buyers and sellers"
- The industry has grown on the supply and demand side.
2. "The products offered by each seller are close substitutes for the products offered by other firms, and each firm tries to make its product look different"
- Most of the performance gear are perfect substitutes except for niche markets.
3. "Entry into and exit from the market are possible with fairly low cost"
- Most firms import from manufacturers abroad and handle the distribution and marketing. It is a low-capital extensive industry.
4. "Firms have some pricing power"
- This has been declining. This decline is due to factors such as excess inventory and the increase in competitors.
5. " Suppliers differentiate their products through advertising and other non-price strategies"
- Industry incumbents rely a great deal on brand equity. This industry deals with fashion preference, so differentiating through non-price strategies is high relative to other industries.
As the industry has moved closer to monopolistic competition, it has also entered the shakeout phase of the industry life cycle. In this phase, some competitors will be acquired, and some will fail. Price cutting often occurs. A business with a differentiation strategy lacking attributes to differentiate is left in a vulnerable position. Incumbents with greater economies of scale and economies of scope are much more prepared for battle.
Competition for UA has become intense. The firm's performance apparel, which constitutes 78% of sales, has seen many entrants. UA's compression tops are not immune.
The company's ability to alter the performance apparel market is unequivocal. UA's value proposition is "providing superior alternatives to traditional athletic products." Target markets like high school football have found they prefer the moisture wicking fabrications over cotton. The question we postulate is how the market will evolve for companies like Under Armour, which utilize a differentiation strategy, where patents only protect brand equity?
Michael Porter's five forces is a tool to understand the pressure a firm faces in the competitive environment. A statement from the companies 10- K explains it explicitly: "Our industry is subject to significant pricing pressure caused by many factors, including intense competition, consolidation in the retail industry, pressure from retailers to reduce the costs of products and changes in consumer demand. These factors may cause us to reduce our prices to retailers and consumers."
The firms other products such as footwear and accessories are competing in a more established arena. While both segments have added to net sales, its unlikely that these products will continue a large growth trend. Incumbents like Nike and Adidas have been producing similar goods for a while.
4. Corporate Strategy
UA has three foreseeable choices given the environment; two of which provide a conundrum for shareholders. UA can keep its pricing strategy by accepting a lower market share, or it can lower its pricing to maintain or increase its market share. The best option is to introduce new products to market, and continually differentiate. This is very difficult to do. We find it unlikely that UA will be able to provide this long term against very capable incumbents. As Eli Portnoy has noted, " It appears however that Under Armour has actually introduced more product that is similar to existing offerings from Nike and others outside the undergarment segment trying to leverage the success of the brand into more traditional athletic apparel."
Sales to women are one place for growth, but unlikely to occur. Sales by gender are not categorized in the company's SEC filings, therefore we can not provide a trend. It is reasonable to assume that differentiating by brand image for women will be a competitive disadvantage for UA. UA's "We will protect this house" marketing campaign is unlikely to attract the woman performance apparel market. Lululemon and Nike should continue to win in this space.
International sales is the other area for significant growth. The company provides information on it's geographical sales. The results are not extraordinary. On page 4 of the 2011 10K, it shows that sales to "other foreign countries" were 5.7%, 6.2%, and 6.1% through 2009, 2010, and 2011, respectively. The Company has been trying to increase its brand recognition by sponsoring teams primarily in rugby and soccer.
In July 2006, UA released information on distribution agreements it made over Europe and the Pacific Islands. The agreements were with six distributors and master agents. Three of these were Equation Performance in France, Main Sport in Germany, and Sportbox SRL in Italy. This information argues that UA is unlikely to become a dominant player oversees, given its inability to grow a posteriori. On the basis of ex ante, incumbents like Nike with strong international distribution and brand equity, along with competitors such as Skins, will make it very difficult for UA to achieve significant international market share.
UA's intellectual property is discussed on page 7 of its 10K with the most important paragraph being at the bottom of the page. It states: "The intellectual property rights in much of the technology, materials and processes used to manufacture our products are often owned or controlled by our suppliers." Under Armour's inability to control these patents have, and will continue to assert pressure in three of the five areas in Porters.
The company released Charged Cotton in 2011. This is a natural cotton that is supposed to perform like UA's synthetic brands. We are unsure of this move. It may provide leveraging of brand equity, however, it could dilute it. It appears more defensive than offensive, and puts the firm's value proposition in question. Regardless, Charged Cotton does not provide the blue-ocean strategy its shareholders need.
We are not optimistic toward the companies long term ability to provide a competitive advantage and achieve the growth that is currently priced into the stock. As we mentioned in the Investment thesis section, this may be the result of a negative cognitive bias. In response, we believe our points hold strong logic. Never the less, as the valuation section will describe, we model share price inconsistent to our pessimistic views, and assume significant market share growth in each scenario.
To arrive at a per share value for UA, we use the free cash flow to the firm approach (FCFF). FCFF is the amount of cash a firm generates to shareholders after operating expenses, interest, principal payments, capital expenditures (capex), and net working capital (NWC) have been made. The FCFF equation:
FCFF = Net Income + NonCash Charges + Interest Expense*(1-Tax Rate)-Capex - NWC
To forecast the cash flows we used a three-period model. The first period, or excess growth period, lasts for five years (2012-2016). The second period, the slowed growth phase, lasts for four years (2017-2020). Lastly, the terminal phase is the projected cash flows from UA into perpetuity, or as we shall call it, terminal value. The projected FCF's for each year are discounted at the weighted average cost of capital (WACC) and summed. We add back cash and subtract long-term debt to arrive at value to common equity. This value is divided by the number of shares outstanding to get a per share value.
Assumptions: The assumption making that utilizes imperfect information is meant to err toward overvaluation.
- Risk-free rate (RF): we use a 4% RF starting out then add 25 basis points each year. This attempts to capture our belief that interest rates will rise during the span of the projected cash flows.
- Risk premium (RP): 4%, which is well below the historical average
- Cost of debt: 3.5%
- Growth rate of the economy: 3.5%
- Tax rate: 38%
- Profit margin: 7%; margin is higher than what has been achieved in the past few years.
- Current share price: $58.20 as of the close on Aug. 31, 2012
- Shares outstanding: 104,460,000 taken from most recent 10-K
WACC is the cost of equity and debt weighted to the company's capital structure. This is what we will use to discount all of UA's projected cash flows. The formula for WACC:
WACC = (D/V)*Cost of debt *(1-Tax rate) +(E/V)*Cost of equity
UA's capital structure is very one-sided with low levels of debt. This may look like a positive attribute, but as we will convey, it puts a hindrance on its returns. D/V and E/V are the percentages of debt and equity in the capital structure and are computed using market values (MV). MV of equity equals $6.08 billion, MV of debt equals $106 million, and therefore E/V and D/V equal 98% and 2%, respectively.
To calculate cost of equity we first need to calculate beta of equity for UA. This measures the return that shareholders demand on UA stock. Since UA is a young company and has not been public for that long, we wanted to use a comparable firm's beta and average the two together. The comparable we chose is Nike (NKE). Both UA and Nike's betas were estimated using return data from 2005. UA and Nike's return series were regressed against the S&P 500 for the same timeframe and produced a beta of 1.45 and 0.83, respectively. The two betas were averaged together to get a beta of equity of 1.14, which we believe is optimistic given the large dispersion between the two. This produces a WACC of 8.45% for 2012.
Valuation (U.S. Scenario Emphasis)
The multiple scenarios we use for valuation all follow the same process and use the same model. To make the following sections easier to comprehend, we use the US scenario to explain the valuation process. The following explains our figures for the FCFF model.
For the U.S. case, we made the assumption that growth would continue strongly in the US, but be insignificant internationally. As you can see below, the assumption for market growth was 2.5%. UA currently holds approximately 3.2% market share. We assume market share will double in five years and triple in ten years. With these optimistic market share assumptions, revenue still falls short of what analysts are projecting.
Net income was forecast using an average profit margin for the past six years and comes out to 7%. We feel this is optimistic due to UA's place in the industry life cycle, level of competition, and volatile margins in the past.
Non-cash charges (NCC) are charges that reduce income and do not have a cash outlay. NCC was forecast for each scenario using a number of techniques. Balance sheet items were forecast using the percentage of sales or cost of goods sold method and use the respective scenario's revenue forecasts. The balance sheet forecasts determine what the NCC will be for each year.
Interest expense remains the same through 2015 when its debt comes to maturity. Due to our inability to predict UA's debt/equity, we have interest expense being zero after 2015 for all scenarios. We will explain later how this might be an opportunity to increase its cash flows and lever its returns. The company states in its 2011 10K that capex will be $65 million in 2012. After 2012 we model capex growing at 5% per year through the excess growth period and then decreasing by 5% during the slowed growth phase.
Net working capital (NWC) is accounts receivable (A/R) plus inventory minus accounts payable (A/P). A FCFF model uses the change in NWC (△NWC) which is the current year's NWC minus the previous year. The model uses the forecast balance sheet and cash flow statement items to determine NWC. NWC grows each year at a steady pace which is expected due to UA's place in the industry life cycle.
When valuing an asset it is best to use a discount rate that reflects the dynamic nature of interest rates. We model that the risk free rate will increase by 50 basis points per annum. This protects against the eventual rise in interest rates. By 2020 our model's RF rate is 6%, which is closer to the historical average. WACC is updated on a rolling basis for each year based on the new RF rate.
Terminal value is the third phase in our model and is what we predict the company will grow at into perpetuity and accounts for the bulk of UA's per share value. It is calculated as 2020's FCFF discounted at the projected WACC minus the growth rate of the economy. The equation: terminal value= (FCFF2020)/(WACC 2020-3.5%) Terminal value comes out to be $4.16 billion for the U.S. scenario.
The FCFFs are discounted at their respective WACCs and added together. Cash and equivalents are added and long-term debt is subtracted to get value to common equity. Finally, value to common equity is divided by the number of shares outstanding to get an intrinsic value of $25.08 per share.
Global Scenario and Best Case Scenario
Global Scenario and Best Case Scenario use the same FCFF methodology, but use different top-line projections. The global scenario is based upon the success of global expansion. It assumes that UA will be able to achieve a certain percentage of the global market share in five and ten years time. Best case is also based on global market share expansion, but with more optimistic revenue projections. Intrinsic value per share for global and best case come out to $24.46 and $28.12, respectively. The top-line projections are outlined below.
For the Global case, we used the global performance apparel market as the overall size of the market, which was approximately $122 billion. The growth rate assumed was 3.2%. UA's current market share is approximately 1.2%. We assume that UA will double market share in five years and triple market share in ten years. Here is the top-line projection with the above inputs:
Click to enlarge images.
For the best-case scenario, market growth is assumed to be 5%. Market share is assumed to be 3% in five years and 5% in 10 years. This means UA will approximately triple its market share over the globe in five years. Given the qualitative and historical data presented earlier in the corporate strategy section, this appears unlikely.
6. Expected Value
In our attempt to not predict the future, we use an expected value approach in projecting share value. Many outcomes may play out for UA. We assigned probabilities to each of the three scenarios we outlined in the valuation section.
7. Closing Thoughts
We have projected a share price 56% less than current market value. Its quite obvious that the market is valuing UA as a high flying growth company. Our thesis is that these projections are unrealistic and unsustainable. UA currently trades at a P/E multiple of over 60 and an implied growth rate upwards of 25%. As is the case with any high P/E stock, either the E as in earnings has to catch up or the the P as in price has to fall. In our case, we believe at best both will move, and at worst, only the price will decline.
When evaluating how well a company operates it is best to go to the cash flow statement. For UA, there are a few conclusions we draw from this. Earnings has trended upward every year from the bottom of the recession, yet cash flow from operations (CFO) has declined each year. This points to a number of issues. One, increasing net income without increasing CFO is a sign of poor earnings quality. Two, CFO should cover the cost of capex, which implies that the company generates FCF.
Going back all the way to 2006, only in 2008 and 2009 did UA have enough FCF to cover its capex for the year. For 2012, we have the following FCFF projections:
UA says that capex will be 65 million for 2012 and we have shown that it can barely cover this under the U.S. and global scenario, while for best case it falls short substantially. FCFF for 2012 under the best-case scenario is low because of the aggressive growth assumptions that the model uses. In the best case, items under current assets grow at a faster rate than the other scenarios for 2012.
UA often understates its level of debt, and books gains it should not. This points to a lack of integrity, and raises the question: Why does a company with such amazing growth prospects need to utilize aggressive accounting policies?
UA extensively uses operating leases for its retail locations, offices, and warehouses. When a company uses operating over capital leases, profitability, ROA, and debt ratios can appear to be higher than they should. The leased asset is not put on its balance sheet as an asset because it does not own the leased asset. All UA does is record rent expense and put it in SG&A.
UA makes use of a revolving line of credit for $325 million and uses it for working capital and other financing needs. It does tend to make use of the line of credit, but always close it out before years end. When looking at current liabilities you will notice an item called "current maturities of long term debt." We believe UA classifies long-term debt as current liabilities to understate how much its debt has actually increased. This gives the effect of a stronger balance sheet. In our analysis, we did add back this item to get long-term debt. UA is going to have a hard time paying off its current liabilities with its paltry amount of cash coming in at $107 million. Its cash level has decreased significantly from 175 million at the end of 2011. UA may need to tap its line of credit or issue new equity to cover these expenses.
The effort of management to keep debt levels artificially low may actually put a hindrance on its returns. As we have shown in the WACC calculation, 98% is weighted on equity. If it was to take on additional debt, which it may need to do, WACC would decrease. Taking on additional debt is not a bad thing when its cost of debt is so low. Though it may have trouble making interest payments due to its inconsistent cash flows, it is still something to consider. When discounting cash flows, a lower WACC would lead to a higher per-share value.
In 2011, UA booked a gain on the purchase of land for its corporate headquarters. It paid $60.5 million, but put fair value on the land at $63.8 million giving them a gain of $3.3 million. UA sites the reason being that the seller did not take proper due diligence in advertising the land properly. We find this shady, given that historically information asymmetries exist for the benefit of the seller. To make sure, we looked over UA's 10-K and found no indication that it is a real estate valuation professional. What's worse is that UA booked this "gain" in SG&A expense and classified it as an operating cash flow. UA is not in the business of buying and selling land and therefore should not be booking gains as such. This event seems to fit the GAAP standards of unusual in nature and infrequent in occurrence; deeming it extraordinary.
The operating income target for 2011 on the high end is 160 million. UA's operating income for 2011 was $163 million. Coincidentally, earnings were beat by the amount of the real estate gain. When looking at additional proxy statements, we found that UA frequently mentions how it beat its target level. This is a recent statement from a 14A proxy statement: "Operating income was $163 million, an increase of 45% over 2010, and well above the target for maximum incentive awards under the plan of $160 million." It's surprising that a 1.875% beat on operating income could be considered "well above." Orwellian doublespeak was found in numerous places throughout UA's SEC filings.
|2010 results||2011 targets||2011 results||2011 vs 2010 results|
|Net revenue||1.064 billion||1.2 billion||1.473 billion||38%|
|Operating Inc||112 mm||120-160mm||163mm||45%|
|Op inc %||10.60%||10.5%-11.5%||11.10%||0.005%|
CEO Kevin Plank and other executives have been on a selling spree of late. The selling is a part of the 10b5-1 trading plans, which was enacted by the SEC in 2000. A study in 2006 has shown that despite the efforts to decrease the asset of material nonpublic information for insiders, abnormal returns are taking place through these plans. We are skeptical of UA's insider selling and perceive it as bearish news, regardless of the 10b5-1 trading plan.
When valuing any company, the quality of management must be considered. Plank owns 73% of the voting interest in UA. Plank is the founder of UA, so a large voting control can be expected. What this means for investors, though, is that it will be hard to hold executives accountable for their actions. Good luck to investors trying to make change in a company where the majority rule is owned by a CEO who lets the things we have mentioned pass his judgment.
Currently, UA sports a market cap of over $6 billion. In light of the other information we have presented,the valuation appears very irrational. With our valuation of $25.69 per share, UA's market cap would be $2.7 billion. We project the forward P/E for 2012 to be 20.8 compared to analyst consensus of 38.73.
The market is pricing in unrealistic growth, idealistic earnings quality, and turning its eye to insider selling. "Investors may at times be lured into making overly optimistic projections based on temporarily robust results, thereby causing them to overpay for mediocre businesses." With the stock trading at approximately $59, UA presents a good short opportunity. Greater than 50% margin of safety should provide significant room for error in the analysis.
(Note: Anyone interested in obtaining the spreadsheet used in the analysis can email me.)