Case UA Q+A 20240410

1 Answers CASE = UNDER ARMOUR INC. (UA) Version: 4/10/24 This case has you analyze the financial statements of, and disclosures made by, Under Armour (UA), especially with regard to UA’s revenues during the period Q3:2015-Q4:2016. At the top of the ocean, you’ll need to critically appraise the motives, decisions and actions of the key players in the case. Down in the krill, you ’ll need to take the information in the Case and other documents on Canvas plus the incomplete data in the case Excel file on Canvas, find missing P&L data from SEC.gov and then do a number of calculations, analyses and evaluations. Balance the two perspectives well, and good luck! Q0. Who is the key FSAV-oriented decision maker in the UA case? What business decision is he faced with? How could high-quality FSAV improve his decision? (1.5 pts) DM = Scott Baxter, a recent MBA from the Ivey Business School at the University of Western Ontario. It’s May 2021 and Baxter is monitoring his personal investment portfolio, part of which is long in UA. He sees the news headline “UA to Pay $9 million to Settle SEC Charges” and is surprised to see only a slight drop in UA’s share price in response to this announcement. The business decision that Baxter faces is: What should he do with his long position in UA? Should he sell some or all of his shares? Or none? Should he buy more? Or go short? High- quality FSAV could improve his decision by giving him a sophisticated investor’s understanding of the “what/when/why/how/who/past -present- future” behind the SEC’s charges against UA + settlement. A sophisticated investor will be one who understands what happened, when it happened, why it happened, how it happened, who made it happen, and what the implications are for UA’s future. A sophisticated investor will also differentiate out economic aspects from accounting aspects; within economic aspects, fundamental from capital market; within accounting aspects, recognition from disclosure. Lastly, a sophisticated investor will also appreciate the viewpoints on the “what/when/why/how/who/past -present- future” that an unsophisticated investor has, since the probability that the marginal investor setting UA’s stock price is sophisticated is less than one, not only now but in the future. ___________________________________________________________________________________ Q1. Who else might do FSAV on UA over the 5 years leading up to May 2021, and why? (1 pt) Here are the most likely candidates and reasons: Equity analysts What do I forecast UA’s REV, EPS etc. will be over the next Qs/Ys? What do I forecast UA’s stock price will be over the next Qs/Ys? Individual investors like Baxter For same reasons as Baxter in Q0. Specialized/fraud researchers Is fraud going on given how fast REV are growing? Competitors (Nike, Adidas) How is UA achieving such consistently high REV growth? How big of a competitive threat is UA to us? How do we respond? UA CEO Plank + UA CFO What do I forecast UA’s REV, EPS etc. will be over the next Qs/Ys? Hedge funds Is UA overvalued or undervalued? SEC, Justice Dept. Is there fire in UA REV smoke concerns that have been brought to us? Class-action lawyers Has UA or key individuals at UA acted illegally? Big customers of UA Why is UA asking us to push more & more REV forward? WSJ, Bloomberg, Jim Cramer Is there a story here that our readers would be interested in? Credit rating agencies Is UA creating credit risks because of its rapid REV growth strategy? Politicians like Sen. Warren Are execs abusing 10b5-1 Should there be more Govt. regulation?

2 Q2. Using the horizontal timeline below, visually depict around and connecting to the timeline the major events in the case. You may draw freehand or use software. Your timeline does n’ t need to be linear — so 1 cm can separate 2 events that are one week apart or one year apart. You may use abbrevs. I recommend that you use [1] arrows, [2] boxes or circles with text in them, [3] coloring them differently based on what type of event it is. NOTE: I ’ ll grade generously because I’m not looking for you to get exactly the same timeline as I do – just some of the key events. (2 pts) FYI: I ’ ve added 2 relevant events that are not in the case. The first is the 10/15/15 announcement that CFO Brad Dickerson was leaving, and ex-Navy Top Gun fighter pilot Chip Molloy was replacing him. Maybe Brad decided to bail from UA’s hyper -aggressive REV approach, but per the associated press release, Top Gun Molloy was eager to take on the risks. Then after the SEC REV debacle, Molloy decided to leave for “personal reasons” and was replaced by someone who, from the associated press release, sounds like “just a regular normal guy”.  “Over next decade (06 -15), UA continued → to see strong financial performance, primarily driven by nearly 20% quarterly REV growth” (p.2)  High competition, pessimism → + Sports Authority liquidates (p.2) Q3 2015 Q4 2015 Q1 2016 Q2 2016 Q3 2016 Q4 2016 Q1 2017  Plank 1 st 10b5-1 → stock sales ($458 mil) First ever Q loss Oct 2019 Plank to step down from CEO on 1/1/20 Plank boasts of 20%+ Q-over-Q REV growth past 5 years  Plank 2 nd 10b5-1 → stock sales ($138 mil) 1 st CAL filed 1 st CAL dismissed 1 st CAL reopened WSJ: UA is being probed by DoJ & SEC for pulling REV forward 2020 11/4 2019 UA reveals it has a Wells Notice from SEC 7/27 2020 5/3 2021 SEC press release: UA charged with failing to disclose $408 mil of RPF + case settled for $9 mil. + no restatements UA stock hits all-time high ≈ $50 and mkt cap ≈ $50 bn UA stock falls 58% 2017 COVID + lower REV guidance + restructuring  SEC’s “Relevant Period” per Wells Notice → 6 Quarters during which SEC alleged UA [1] pulled forward $408 million of REV to help meet-or- beat analysts’ REV estimates, but [2] materially misled investors by failing to disclose to them the impacts of the RPFs on future REV. UA IPO; 1 st day pop = 92% 11/18 2005 Sept 1996 Plank founds UA CFO Brad Dickerson announces he’s leaving. New CFO is Chip Molloy, an ex-Navy Top Gun fighter pilot 2011 10/15 2015 CFO Chip Molloy announces he’s leaving “for personal reasons”. David Bergman . Previously SVP + with UA since 2004. 1/31 2017

4 in my experience, the safest place to find missing P&L data, and where Baxter would most likely go given that he’s no longer enrolled at Ivey B. School, is www.sec.gov . There, follow these steps: > FILINGS / Company Filing Search / “Under Armour” / [+] 10 -K (annual reports) and 10-Q (quarterly reports) / View all 10-Ks and 10-s / pull_down_appropriate_10-K_or10-Q in Form description column / scroll thru until you find the correct annual or quarterly P&L / fill missing data into Excel One wrinkle is that separate Q4 data is not reported in a firm’s 10 -K. But you can calculate say Q4 REV from the full year’s REV less the sum of quarterly Q1+Q2+Q3 REV. Once you have all the annual and quarterly REV data nailed down, it’s straightforward in Excel to create the graphs. It’s also straightforward to fit a trendline by right-clicking on (here the orange) line / Add trendline . You can then choose from a variety of Trendline Options = Exponential / Linear / Logarithmic / Polynomial / Power / Moving Average. For both REV series, my eyeball-test suggests that a 3 rd -Order Polynomial provides “the best fit” trendline, as it captures the key feature that REV is increasing but at a decreasing rate . Importantly, a linear fit is NOT the or a “best fit” because even by eyeball, neither UA’s annual nor quarterly REV is linear in time. Instead, it is non-linear. A linear fit will be mis specified in several ways, including but not limited to the differences between actual and fitted REV being autocorrelated. Also importantly, it is not just the Statistics angle that speaks to and declares that a 3 rd -order polynomial is the best fit for UA’s REVs. Economics predicts a 3 rd -order polynomial because the 1 st derivative of a 3 rd -order polynomial is a 2 nd -order polynomial, and a 2 nd -order polynomial nearly captures the economic reality that for UA, “REV is increasing but at a decreasing rate” . What’s SIMILAR What’s DISSIMILAR • Rapid growth end of 2016. Then growth markedly slows down. • Seasonality in Quarterly REV: Q4 > Q3 > Q1 > Q2 (due to many sports really taking off in the Fall) • You see COVID nailing both quarterly and annual REV in 2020. • More hints of REV growth having entirely flattened out / stopped as of about Q3:2016 in quarterly REV than in annual REV? • Best Excel trendline is 3 rd -order polynomial. $(1,000,000) $- $1,000,000 $2,000,000 $3,000,000 $4,000,000 $5,000,000 $6,000,000 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020` 2021 UA reported annual REV ($000s) Annual REV Poly. (Annual REV) $- $200,000 $400,000 $600,000 $800,000 $1,000,000 $1,200,000 $1,400,000 $1,600,000 $1,800,000 Q1:2005 Q3:2005 Q1:2006 Q3:2006 Q1:2007 Q3:2007 Q1:2008 Q3:2008 Q1:2009 Q3:2009 Q1:2010 Q3:2010 Q1:2011 Q3:2011 Q1:2012 Q3:2012 Q1:2013 Q3:2013 Q1:2014 Q3:2014 Q1:2015 Q3:2015 Q1:2016 Q3:2016 Q1:2017 Q3:2017 Q1:2018 Q3:2018 Q1:2019 Q3:2019 Q1:2020 Q3:2020 Q1:2021 Q3:2021 UA reported quarterly REV ($000s) Quarterly Net Revenues Poly. (Quarterly Net Revenues)

5 Q5a. The case + the SEC’s AAER use these three ambiguous phrases to describe UA’s REV growth: “quarter -over- quarter REV growth” (case p.2), “quarterly REV growth” (case p.2), “year -over- year REV growth” (SEC p.1). Provide 3 different algebraic definitions of % REV growth D1, D2 and D3 that taken together as a set cover/span what could plausibly be meant by these ambiguous phrases, denoting REV t-k = $ revenue in quarter t-k. (1.5 pts) D1. (REV t – REV t-4 ) / REV t-4 D2. (REV t – REV t-1 ) / REV t-1 D3. ( ∑ REV 𝑡−𝑘 3 𝑘=0 – ∑ REV 𝑡−𝑘 7 𝑘=4 ) / ∑ REV 𝑡−𝑘 7 𝑘=4 ) [based on trailing 4 quarters of REV; 1 st best] or (AREV t – AREV t-1 ) / AREV t-1 [2 nd best], where AREV t {AREV t-1 } = REV over the most recent fiscal year {the fiscal year immediately before the most recent fiscal year}. I ’ ll also accept D4 = a continuously compounded measure of quarterly or annual REV growth. ___________________________________________________________________________________ Q5b. About UA’s REV growth, p.2 of the case states that “Over the next decade, UA continued to see strong financial performance, driven by nearly 20% quarterly REV growth.” By doing your own calculations, determine if the case writer’s statement is True or False. (1 pt) False. Assuming that at its Nov. 2005 IPO, “ the next decade ” means the 40 quarters Q1:2006 → Q4:2015, then my calculations show that none of D1, D2 nor D3 yields an average close to 20%. D1. (REV t – REV t-4 ) / REV t-4 = 31% D2. (REV t – REV t-1 ) / REV t-1 = 10% D3. ( ∑ REV 𝑡−𝑘 3 𝑘=0 – ∑ REV 𝑡−𝑘 7 𝑘=4 ) / ∑ REV 𝑡−𝑘 7 𝑘=4 ) = 30% Q5c. On p.2 of the case, the case writer states “In October 2015, CEO Plank had boasted about the performance of UA on an earnings call with Wall Street investors and analysts. Plank focused on UA’s 20% quarter -over-quarter REV growth for the past five ye ars …” By doing your own calculations, determine if the case writer’s statement is True or False. (1 pt) False. Assuming that as of Oct. 22, 2015 (the date of Q3:2015 earnings call per the call transcript) the "past 5 years" means the 20 quarters Q4:2010 → Q3:2015, then my calculations indicate that none of D1, D2 nor D3 yields an average close to 20%. D1. (REV t – REV t-4 ) / REV t-4 = 30% D2. (REV t – REV t-1 ) / REV t-1 = 10% D3. ( ∑ REV 𝑡−𝑘 3 𝑘=0 – ∑ REV 𝑡−𝑘 7 𝑘=4 ) / ∑ REV 𝑡−𝑘 7 𝑘=4 ) = 30% -30% -20% -10% 0% 10% 20% 30% 40% 50% 60% 70% Q1:2006 Q2: 2006 Q3:2006 Q4:2006 Q1:2007 Q2: 2007 Q3:2007 Q4:2007 Q1:2008 Q2: 2008 Q3:2008 Q4:2008 Q1:2009 Q2: 2009 Q3:2009 Q4:2009 Q1:2010 Q2: 2010 Q3:2010 Q4:2010 Q1:2011 Q2: 2011 Q3:2011 Q4:2011 Q1:2012 Q2: 2012 Q3:2012 Q4:2012 Q1:2013 Q2: 2013 Q3:2013 Q4:2013 Q1:2014 Q2: 2014 Q3:2014 Q4:2014 Q1:2015 Q2: 2015 Q3:2015 Q4: 2015 Q5b: UA Reported REV growth (40Qs Q1:2006 - Q4:2015) D1 D2 D3

7 The value of taking a “ nullius in verba ” attitude to UA’s REV is high for these reasons, I believe: First, if you take someone’s word for how large (or small) a firm’s REV growth is, you risk being wrong. Second, the more credible/expert is the source, the more credible should be their measure of REV growth. This is evidenced by the SEC’s statement being the correct one given that the SEC is staffed by experts. Third, when the case writer states that “In October 2015, CEO Plank had boasted about the performance of UA on an earnings call with Wall Street investors and analysts. Plank focused on UA’s 20% quarter - over- quarter REV growth for the past five years …” , it would seem that either Plank was misinformed by his CFO, or he was being strategically or deceptively conservative, because both D1 and D3 definitions of growth in reported REV are 30%, assuming that the “prior 5 years” = 20 Quarters Q4:2010 → Q3:2015. Fourth, however, it turns out that Plank wasn’t misinformed by his CFO, nor was he being strategically or deceptively conservative. Rather, the case writer has incorrectly quoted what Plank says in his opening remarks in UA’s Q3:2015 earnings call ! Nowhere in UA’s Q3:2015 earnings call transcript does Plank or his CFO talk about REV or REV growth “for the past five years” . [Confirm this by searching the file UA_3Q15_Transcript for the word “five”]. W hat Plank actually says about UA’s past REV growth is that [1] UA has recorded “22 consecutive quarters of 20 -plus percent REV growth” (p.3), and [2] “Our 10 years now of … 30% top-line … growth …” With regard to [1], y es, 22 quarters is close to being five years = 20 quarters. But it isn’t the same. More importantly, the case writer left off what I’ve highlighted in bold green , namely the critical text “ -plus ”. The “ -plus ” is key because g oing by REV growth definition D1, Plank is indeed correct – over the prior 22 Quarters Q2:2010 → Q3:2015, REV growth per D1 had been 20% or more. [Side note: Not so for D3, however, in that D3 REV growth in Q2:2010 was 19%. And certainly not for D2 REV growth definition]. With regard to [2], Plank is correct. Both D1 and D3 definitions of REV growth come in at about 30%. ___________________________________________________________________________________ Q6. According to the case and SEC AAER 4220, what real action(s) and what real inaction(s) did UA take with regard to its REV during the six-quarter window Q3:2015 → Q4:2016 that led to the SEC issuing AAER 4220 against UA? Were any of them in violation of US GAAP? Were any of them in violation of the US Securities or Exchange Acts? (2 pts) Real actions: UA used REV pull forwards to help it meet or beat analysts’ REV forecasts. BUT … Note that p ulling REV forward and booking such REV per se does / did not go against US GAAP as long as the firm has met all the conditions for recognizing REV, which UA did. UA’s pulling of REV forward was therefore per se not a violation of US GAAP accounting REV recognition, nor was it a violation of the US Securities or Exchange Acts. Indeed, at the bottom of p.4 of the case, it is stated that “The Form 8 - K (relating to the Wells Notice) … stated that “the SEC Staff has not alleged any revenue re cognition or other violations of GAAP …”” In particular, the SEC did not allege that UA had been channel-stuffing. Real in actions: UA misled investors by making positive statements about its revenue growth rate and the factors contributing to the revenue growth rate, without disclosing the significant impact on revenue from its use of pull forwards .

8 In theological language , UA’s sin was one of omission more than commission. Yes, on the commission front UA had pulled a material amount of REV forward – indeed, enough REV that in each of the 5 quarters Q3:2015 → Q3:2016, REV including the pull forwards was such that it met or beat analysts’ expectations/forecasts of what REV would be. But it was on the omission front that UA went off the rails in that the SEC alleged that by not disclosing its material REV pull forwards, UA had [1] misled investors as to the basis of its REV growth in the quarter in which the pulled-forward REV was booked, and [2] misled investors as to the persistency of its REV growth because by not telling investors that some of the current quarter REV was pulled forward, investors did not know to revise downwards their REV forecasts for the next quarter (out of which the REV had been pulled).

10 The key takeaway from the True REV numbers is that for each quarter Q3:2015 → Q3:2016, by pulling REV forward one quarter, UA turned what would have been a REV miss into a REV beat. NOTE** T he method I’ve taken above is what I would call the “partial derivative” approach , in that I deliberately stated in the wording of 3. with regard to defining True REV that True REV was defined as REV if UA had pulled no REV forward into the quarter , all else held equal. The emphasis being on the “all else held equal” part because I think it’s plausible to assume that UA made its decision about whether to pull any REV forward and if so, how much REV to pull forward on a quarter-by-quarter basis. Under this presumption, the fact that REV in the current quarter is lower than it otherwise would be by the amount of REV pulled forward into the previous quarter is irrelevant because it is given and can’t be changed by UA in the current quarter. However, there is also another method that could be used to estimate True REV, which I’ll call the “total derivative” approach. In this approach, you assume that True REV is what REV would have been if UA had never pulled any REV forward in any quarter, period. Under this presumption, the fact that REV in the current quarter is lower than it otherwise would be by the amount of REV pulled forward into the previous quarter is NOT irrelevant – it definitely has to be taken into account! I show below what True REV would be under the total derivative approach. The key difference is that under the total derivative approach, True REV sometimes misses and sometimes beats analysts’ consensus REV forecasts , whereas under the partial derivative approach, True REV always misses analysts’ consensus REV forecasts. This contrast suggests that UA did indeed separately decide each quarter whether to and how much REV to pull forward. Step Q3:2015 Q4: 2015 Q1:2016 Q2:2016 Q3:2016 Q4:2016 1. Quarterly Reported Net Revenues ($000s) REV 1,204,109 $ 1,170,686 $ 1,047,702 $ 1,000,783 $ 1,471,573 $ 1,305,277 $ 2a. Loss of REV Pulled Forward from prior Q RPF_LPQ - $ (45,000) $ (99,000) $ (17,500) $ (10,000) $ (65,000) $ 2b. REV Pulled Forward into the Q (per AAER) RPF 45,000 $ 99,000 $ 17,500 $ 10,000 $ 65,000 $ 172,000 $ 3. REV if no Rev was Pulled Forward into the Q TRUE_REV 1,159,109 $ 1,116,686 $ 1,129,202 $ 1,008,283 $ 1,416,573 $ 1,198,277 $ 5. Analyst REV consensus ANALYST_REV 1,175,109 $ 1,117,686 $ 1,035,702 $ 1,000,783 $ 1,453,573 $ 1,404,942 $ 4. False REV meet-or-beat F_MOB_REV 29,000 $ 53,000 $ 12,000 $ - $ 18,000 $ (99,665) $ 6. False Rev meet-or-beat (% of AN_REV) F_MOB_REV% 2% 5% 1% 0% 1% -93% 7. True REV meet-or-beat T_MOB_REV (16,000) $ (1,000) $ 93,500 $ 7,500 $ (37,000) $ (206,665) $ 8. True Rev meet-or-beat (% of AN_REV) T_MOB_REV% -1% -0.1% 9% 1% -3% -15% UA's quarterly REV if UA had pulled no REV forward from the next quarter (under the "never any RPFs" counterfactual) Window during which SEC charges that UA pulled forward total of $408 million of REV

11 Q8. All told, if you were the decision-maker, what business decision would you make and why? (1 pt) Recall the business decision that the decision-maker Scott Baxter faces: What should he do with his long position in UA? Should he sell none or some or all of his shares? Or should he buy more? Or go short? Assuming that the US stock market is efficient, which economic intuition + most empirical evidence suggests it is with respect to the situation that Baxter faces, all we’ve covered in the case will have been impounded into UA’s stock price as of the date of the case = right after the announcement on 5/3/21 that UA had agreed to pay $9 million to settle the charges the SEC brought against it. As such, at the top of the ocean, there’s going to be no positive (or negative) expected risk-adjusted return to or in UA’s stock that is available to Baxter based on any of the FSAV that we’ve done! On the one hand this conclusion seems highly dispiriting – Baxter does a ton of FSAV but there’s nothing in it for him? YES – both the blessing and the curse of US equity markets being highly efficient is that the expected risk-adjusted return to public information is zero . This applies to Baxter because all his FSAV was done using public information. None of his information was private / non-public. However, there are ways in which the FSAV Baxter has done can or should help him make a better business decision. Here are some that come to my mind: 1. Baxter slaps himself over the head and says “Dang it! Painful though it is for me to admit it to myself after having done all this FSAV, now I remember what my finance professors taught me about stock market efficiency in my MBA program at Ivey. Going fo rward, therefore, I’m not going to waste my time doing FSAV as a small, part- time retail investor. Moreover, I’m going to stop exposing myself to idiosyncratic risk for no likely gain by holding single stocks such as UA. I’ll get out of my long position in UA and other single stocks and put all my investment dollars into a US market ETF such as the Vanguard S&P 500 Index ETF (VOO) or the Vanguard Total Stock Market Index ETF (VTI).” 2. Baxter’s extensive FSAV gives him a better sense of the SSP sizes, sources and persistences of the fundamental riskiness of UA. That is, FSAV helps him identify and measure the sources, signs and magnitudes of UA’s fundamental “risk factors”, the non -diversifiable ones of which are likely to translate into UA’s stock return “risk factors”. Even if he chooses to go forward still holding UA as a single stock part of his overall investment portfolio, he does so in a more fully informed manner. 3. In doing FSAV, Baxter cares about more than just the financial risk and return to holding UA stock. Maybe he really enjoys doing FSAV and that makes it all worthwhile. After all, para. 1 of the case says that he primarily invested in “the shares of brands he knew and loved” and “he was a proud shareholder in UA”. Or maybe doing FSAV on UA helps him make better business decisions about stocks other than UA. Or maybe holding UA happily reminds him of his younger days when he “played a variety of sports growing up”. There are lots of reasons for doing things that cost you financially but bless you or make you happy!

13 B2. Based on your own analysis of Plank’s stock sales in Exhibit 3, was Plank abusing SEC Rule 10b5- 1 via the scheduled stock selling plan that he entered into shortly after UA’s October 2015 conference call? (2 pts) I do not think the data shown above supports the proposition that Plank abused SED Rule 10b5-1 via the scheduled stock selling plan that he entered into shortly after UA’s October 2015 conference call. During Q4:2015 → Q4:2016, Plank sold at a monthly total-$-value rate of $27.6 million per quarter. But this is only slightly higher (and certainly not statistically significantly higher) than the $24.1 million rate that he sold at during the Q1:2011 → Q3:2015 period. Avg # shs sold per Q # shares sold (mil) $ value of shares sold ($ mil) Approx avg. UA stock price ($) <or interpolated> Q1:2011 2.3 16.0 $ 6.96 $ Q2:2011 2.3 18.0 $ 7.83 $ Q3:2011 2.3 20.0 $ 8.70 $ Q4:2011 2.3 24.0 $ 10.43 $ Q1:2012 1.1 11.0 $ 10.00 $ Q2:2012 1.1 14.0 $ 12.73 $ Q3:2012 1.1 15.0 $ 13.64 $ Q4:2012 1.1 14.0 $ 12.73 $ Q1:2013 1.2 15.0 $ 12.50 $ Q2:2013 1.2 17.0 $ 14.17 $ Q3:2013 1.2 20.0 $ 16.67 $ Q4:2013 1.2 24.0 $ 20.00 $ Q1:2014 1.15 30.0 $ 26.09 $ Q2:2014 1.1 27.0 $ 24.55 $ Q3:2014 1.95 67.0 $ 34.36 $ Q4:2014 1.95 63.0 $ 32.31 $ Q1:2015 1 30.0 $ 30.00 $ Q2:2015 0.95 33.0 $ 34.74 $ Q3:2015 0 - $ 39.11 $ Q4:2015 2.3 100.0 $ 43.48 $ Q1:2016 0 - $ 41.74 $ Q2:2016 0.95 38.0 $ 40.00 $ Q3:2016 0 - $ 35.00 $ Q4:2016 0 - $ 31.00 $ Total Value Sold (TVS) Average TVS per month ($ mil) Q1:2011 --> Q3:2015 458 $ 24.1 $ <-- So evidence is NOT consistent in my mind with Plank exploiting 10b5-1 scheduled stock sales. Q4:2015 --> Q4:2016 138 $ 27.6 $ <-- He sold at monthly rate of $24.1 million in 1st 10b5-1 (Q1:2011 - Q3:2015) vs. $27.6 in 2nd (Q4:2015 - Q4:2016) Exhibit 3 Plank's Share Sales Window 2.3 1.1 1.2 1.5 1.1 0.2 0 0.5 1 1.5 2 2.5 3 $- $10 $20 $30 $40 $50 Q1:2011 Q2:2011 Q3:2011 Q4:2011 Q1:2012 Q2:2012 Q3:2012 Q4:2012 Q1:2013 Q2:2013 Q3:2013 Q4:2013 Q1:2014 Q2:2014 Q3:2014 Q4:2014 Q1:2015 Q2:2015 Q3:2015 Q4:2015 Q1:2016 Q2:2016 Q3:2016 Q4:2016 Approx avg. UA stock price ($) <or interpolated> # shares sold (mil)

14 B3. The chart below displays the cumulative market-adjusted stock returns over the <-10,+10> trading day window surrounding each of three key dates in the case + the SEC’s AAER: [1] UA's Q4:2016 earnings announcement, [2] UA's Q4:2019 earnings announcement [3] UA's disclosure of Wells Notice [4] SEC's release of its AAER 4420 against UA What do you conclude in terms of FSAV from the chart, for each event and comparing across events? State any key assumptions you are making. I assume the US stock market is semi-strong form efficient with respect to public information. 1. UA’s Q4:2016 and Q4:2019 earnings announcements contained really bad news about the present value (V) of UA’s future earnings / free cash flows (FSA). The bad news is most likely to do with the projected levels of UA’s future earnings / free cash flows. But it could be about the riskiness of UA’s fu ture earnings / free cash flows because “growth = risk” and UA is a highly growth -focused company. 2. UA’s disclosure of Wells Notice and the SEC's release of its AAER 4420 against UA contained no news about the present value (V) of UA’s future earnings / free cash flows (FSA). Either the stock market already knew the information contained in these disclosures (there was “no surprise” to be valued), and/or the $ implications of the disclosures were immaterial relative to UA’s market cap. 3. Para 2 on p.5 of the case somewhat misleadingly states that “UA shares sank nearly 4% in pre -market trading after the announcement of the Wells Notice”. Misleading because the 4% drop was very short- lived in that by the end of trading day 7/27/20, UA’s stock price had risen +1.2% (vs. SP500 +0.7%). To me, this indicates that the smart money saw no bad news in the announcement of the Wells Notice. Vertical axis is cumulative market-adjusted returns CMKTADJRET over Event Window <-9, +10>. Market is SP500. The change in CMKTADJRET between the two vertical blue [black] bars = 3-day [1-day] CMKTADJRET that allows for [does not allow for] any pre-ann info leakage and/or post-ann delayed price reaction. -40% -30% -20% -10% 0% 10% 20% 30% <-10> <-9> <-8> <-7> <-6> <-5> <-4> <-3> <-2> <-1> <0> <+1> <+2> <+3> <+4> <+5> <+6> <+7> <+8> <+9> <+10> UA event studies around [1] UA's Q4:2016 earnings announcement (31-Jan-2017) , [2] UA's Q4:2019 earnings announcement (11-Feb-20) , [3] UA's disclosure of Wells Notice (27-Jul-20), [4] SEC's release of its AAER 4420 against UA (3-May-21)