Activision Blizzard is an IP powerhouse unrivaled in the markets it dominates. New titles in the Call of Duty franchise set sales records annually, and secondary revenue sources such as additional content and premium subscription services are boosting RoIC and sustaining revenue well beyond a single holiday season. Every game published by Blizzard is a category killer. World of Warcraft has almost four times more players than the next most popular massively multiplayer online game, and it’s seven years old. Monthly subscriptions provide a steady stream of revenue, bolstered by expansion packs published sporadically. International professional video game leagues have chosen Starcraft as their strategy game of choice for over a decade, and new entrants seem to pose little threat. Starcraft is the 21st century’s chess.
Surprisingly for such a mature business, low-hanging fruit exists to boost sales and margins. The most notable opportunities in this space include “free-to-play” versions of older titles in the Call of Duty series for release in China and other countries with lower GDP per capita. Very low-quality titles in the first person shooter space have demonstrated that this model can be profitable, and Call of Duty titles could easily dominate with a minimal investment in development and marketing. The Blizzard IP extends well beyond just video games. Characters in the Warcraft, Starcraft, and Diablo series have intricate stories and world lore that has been developed over the past two decades through video games, books, board games, and trading card games. The obvious next step is to bring these stories to the silver screen. Development of the first Warcraft movie has been in process for years now, and its success could lead to endless sequels plus development of the Starcraft and Diablo series. On a three-year rotation, (assuming Blizard’s next MMO builds off one of these existing three franchises), the characters and stories can stay fresh. The synergistic effects of marketing and releasing such movies is impossible to predict, but can only have a positive impact on game and merchandise sales.
In the LTM, excluding non-cash impairment charges (I’m not taking out stock-comp), ATVI earned just over $1 billion of net income, or $0.84 per share. However, the release of Starcraft in the LTM inflates revenue and earnings from what we could expect to be a normal run rate, by about $0.20 per share. At $11 per share, this is a trailing multiple of 13.0x (or 17.2x excluding Starcraft). In 2010 and 2009, ATVI repurchased $959 and $1,109 million of stock respectively, over 80% of cash flow from operations. Management was able to do this without throttling growth partly because working capital reductions have contributed meaningfully to operating cash flow. The key contributors have been reductions in AR and inventory caused by a transition from selling boxes through retailers to making games available online for download. This was made clear in a recent investor presentation:
The real risk to the investment thesis is declining revenue from Warcraft. While Activision accounted for 56% of revenue in the LTM, Blizzard accounted for 57% of operating income. In 2010, Warcraft accounted for 89% of Blizzard revenue. So obviously ATVI is extremely dependent on the continued profitability of this one game… which is probably in terminal decline. Modeling the effects of this decline is extremely important. Fortunately, the next Blizzard MMO has been in development for years and could be ready as early as 2013. Even if Titan is launched a year or two late, its official announcement will free analysts to start building Titan revenue into models.
Call of Duty
Activision’s most notable IP is the Call of Duty franchise. CoD dominates the first person shooter market segment, selling twice as many copies as Halo, the next best-seller. This is mostly ecause CoD is available on Xbox, PS3, and PC, whereas Halo is an Xbox exclusive, but CoD outsold Halo even on the Xbox alone. In 2010, CoD was responsible for 31% of ATVI revenue.
Development responsibilities are shared between two companies, Infinity Ward and Treyarch, each with a two-year product cycle resulting in one new game for every holiday season. Ignoring the obvious economic advantages of this arrangement (milking the IP twice as fast), from a risk-diversification point of view, the decision to have two studios work on the project was fortuitous.
Infinity Ward had historically been considered the better of the two development studios, with higher sales for the Modern Warfare spinoffs and better ratings. Treyarch continued to produce games in a WWII setting until Black Ops in the cold war. Immediately following the release of Modern Warfare 2, the executive management of Infinity Ward was fired in a move that shocked the industry. Jason West and Vince Zampella quickly filed suit against Activision demanding back compensation that was withheld. The first sentence of the complaint accuses Activision of “astonishing arrogance” with the second sentence bragging that “Jason West and Vince Zampella are among the most talented and successful videogame developers in the world”. The suit claimed $36 million in damages, although additional claims have since been added.
Activision responded with a complaint of its own, accusing West, Zampella, and Electronic Arts of having “secret meetings” with the intention to “derail Activision’s Call of Duty franchise, disrupt its Infinity Ward development studio, and inflict serious harm on the company”. The suit seeks $400 million in damages and puts forth a strong case that West and Zampella were fired for good cause.
The upside of a win outweighs the downside of a loss, but the most important factor has nothing to do with the lawsuit. Call of Duty Black Ops, the most successful game in the CoD franchise, was produced by Treyarch. Treyarch proved it could produce a competitive offering, and just in time for Infinity Ward to disintegrate:
Despite the problems, the next installment in the CoD Modern Warfare series should be out in time for the 2011 holiday season. It’s been developed collaboratively by a reformed Infinity Ward, a new studio named Sledgehammer, and the Activision internal development team Raven. It’s hard to guess what the ratings (or more importantly the sales) will be, but with the installed base of Xbox and PS3 units rising by an estimated 20%, sales should grow commensurately.
If preorders are any indication, Modern Warfare 3 should easily surpass Black Ops. These charts from vgchartz.com show preorders for both CoD franchises with 13 weeks remaining till launch:
The charts show Black Ops had 504,027 console pre-orders in 2010 (13 weeks out), compared to 1,099,457 for the upcoming CoD title. This also shows that CoD has a healthy lead over Battlefield 3, Electronic Art’s contender praised for a revolutionary graphics engine and destructible environments (two weeks closer to launch). The question then becomes: are pre-orders a good proxy for future sales? This chart compares NA pre-orders for the top games published within the last few years and NA first week sales:
It seems like pre-orders have generally been around half of first week sales and it seems like pre-orders are a decent guide to future sales. Unless the sales multiple decreases significantly from where it’s been historically, CoD is on target to significantly increase sales this year. However, it’s bumping up against the limits of what we could expect sales to be given normal PS3/Xbox attach rates in the 30% range. Normal attach rates would yield 20% growth, in line with installed base growth. For sales to double with a 20% higher installed base, we would need the attach rate to increase to 50%, not a likely scenario. But if the installed base grows by 20% and the attach rate grows from 30% to 35%, unit sales increase 40%, enough to meet my high-end growth estimate for Activision without any help from Activision’s other titles.
The CoD series could also benefit from the re-release of older games in a Free-to-Play format in China and elsewhere. At the moment, a game called Cross Fire is dominating the market, generating revenue in the $1.2 billion range (an estimate I don’t necessarily believe). I haven’t played Cross Fire, and I don’t know if there’s some compelling reason to like it, but the graphics are obviously terrible. Re-releasing an old game to compete with Cross Fire is a move Electronic Arts already made this spring. Activision might have more trouble breaking in now, but it may also be preparing a more compelling offering. At the very least, the move by EA should give some insight into the market opportunity. I haven’t explicitly factored this into my model in any way, the F2P market is pure upside.
Blizzard Entertainment is arguably the most respected game developer in the world. If Blizzard published the next installment in the My Little Pony videogame series, half of the 14 to 25 year-old males in the US with a computer would go out and buy a copy. 1% of them would camp out in tents in front of stores and paint their faces pink with sparkles to pick up their pre-orders. Blizzard games are a must have for gamers.
I linked to this chart above, but I’m reproducing it here as well for full effect with my annotations:
This chart shows how much WoW has dominated the massively multiplayer online (MMO) market and how long it’s held that dominant position. Obviously dozens of other games not represented on this chart are competing for a piece of the MMO pie, but none of these have yet surpassed 1 million subscribers (at least not according to MMOData.net).
But the chart above also shows that Warcraft is probably in decline with little chance of being saved by expansion packs or additional content upgrades. It’s tough to guess what the rate of decline will be, but we can model out some rough scenarios. A base-case scenario is presented below, which I think errs on the conservative side.
The non-GAAP financials provided by ATVI ignore revenue deferment and separate out Blizzard, Activision, and distribution revenue. While my preference is for GAAP, these numbers reveal some interesting information about Blizzards product cycle. Ignoring deferments, Blizzard revenue has looked like this for the past 12 quarters:
In a normal quarter, Warcraft generates about $300 million of revenue and $150 of operating income. Subtracting this out, we can figure that Starcraft generated $451 of revenue and $237 of operating income in its first two quarters. Likewise, the last Warcraft expansion pack yielded $177 million of revenue and $107 of operating income.
My model is driven by a few simple assumptions. I assume that Warcraft will generate $300 million of revenue per quarter, which declines at some rate over the coming five years. I also assume that margins decrease proportionate to revenue declines. As Warcraft subscriptions decline, not only will infrastructure be supported by a smaller player base, but other Blizzard games will become a bigger piece of the pie, indicating that margins will slowly move to be more inline with margins at Activision. I assume margins decline at 1/5 the growth rate, so a 15% decline in Warcraft translates into 3% margin compression. This assumption is arbitrary, but intuitively it looks fair.
On top of this, I add $450 million of revenue from Diablo in 2012, $240 from Warcraft expansions in 2012 and 2014, and then five $25 million payments through 2014 for various Diablo and Starcraft expansions. The timing of these cash flows are based on a leaked production schedule included below, which has so far proven to be fairly reliable:
The excerpt below does not include any revenue from a Warcraft movie or Titan, the next gen MMO. At a 15% decline rate, which I believe is sufficiently pessimistic, my model looks like this:
The purpose of this exercise is really to get comfortable with the four years of declining Warcraft revenues investors might face until Titan. Ignoring all other growth opportunities and the benefits to Activision’s diligent share repurchase program (repurchasing on average 1.5% of the shares outstanding every quarter since 1Q09), the model above estimates a 44% operating income decline from the LTM to 2015. This leaves Warcraft with only 5.3 million subscribers in 2015, still the most popular MMORPG by today’s standards.
As pessimistic as I think I’m being in the model above, there are reasons to be optimistic that Warcraft revenue won’t decline (or decline less). I’m generally not a huge fan of the China growth story, but Warcraft, which is over six years old in the US, is newer in other markets such as China, and its about a year old in Russia. Warcraft has also recently been translated into Portuguese for release in Brazil, another move that can sustain the franchise for another few years. The international appeal of Warcraft should help offset declines in the US, and a movie could revitalize the franchise as well.
The release of Titan, Blizzard’s next MMO, was originally scheduled for 4Q13. A recent interview with Blizzard’s COO indicates that a working version is already running, implying that the game might actually be ready within a year of its original projected release date. Titan developers have also said that they expect Titan and WoW to be able to coexist in the market simultaneously. The sell-side reports I’ve read are not factoring Titan revenue into their models yet. Any new visibility into Titan should increase analyst estimates for 2014 and beyond. I’ve included some Titan revenue in my finished model, with $384 million in 2014 and $768 in 2015. This is the revenue I would expect to see if Titan attracted 4 and 8 million subscribers respectively at the same at the same per month fees that Warcraft players pay. This would make Titan as popular one year after introduction as Aion, an MMORPG produced by the respected Korean studio NCsoft. At a 7% decline rate, this still leaves Blizzard generating less revenue in 2015 than was earned in 2010, so it seems fair.
The CoD franchise’s strongest competitor was Halo, developed at the studio Bungie. After the release of Halo Reach in 2010, Bungie parted ways with Microsoft. Further development of the franchise is left to Microsoft’s in-house studio 343 Industries, established in 2007 specifically to carry the torch. Bungie now has a 10-year distribution agreement with Activision in which Bungie retains ownership of its new IP and Activision has exclusive global distribution rights. With Halo under the stewardship of a new design team, it’s difficult to guess how competitive it will be in future years. It will probably sell extremely well, but a lot is riding on their first installment, and that’s not due out till next year.
While no release dates are set for Bungie to produce its first new IP, the Bungie brand name is well known and any game they produce will be given the attention other similar games might not receive. I believe the distribution rights are valuable, although with no insight into what Bungie is producing, or more importantly, when it might be released, it’s impossible to guess what that value might be. Again, I assume no benefit to the “distribution” revenue line, but upside opportunity exists.
Video games aren’t as subject to fad risk as some investors might think. The top selling franchises have been around for decades, with Call of Duty earning the first spot on Wikipedia’s list for a game created within the last decade. Many games toping this list are more than 20 years old, demonstrating the staying power of established brand names.
Guitar Hero (distributed by Activision) and Rock Band (distributed by Electronic Arts) are notable recent notable failures. Despite the ubiquity of the games, rumors spread that the original game developer of both series, Harmonix, was sold for $49.99… dollars, not millions. The acquirer assumed around $200 million of liabilities. This was the result of a meltdown, not just for a franchise, but for a genre. Financially, the impact to ATVI was significant, including not only lost sales, but millions in asset impairments and write-downs. The loss of this genre has been a drag, contributing to the flat revenue for ATVI and losses for EA.
Part of the blame for this collapse was levied on Activision for “milking” the brand, an accusation that I believe was warranted to some degree. That being said, as an investor, the business decision was sound. Perceiving the demise of the genre, Activision published 7 Hero titles for platforms (not including iPhone or handheld) in 2009, compared to four in 2008, two in 2007, and two in 2010. According to the 2008 annual report, one of Activision’s titles was the first to ever to generate $1 billion of sales (as a single title). Milking the brand on its way out probably recovered as much revenue as possible, and destroying the genre ruined EA’s chances of a big hit with Rock Band III, which was a legitimately innovative offering. While Activision hasn’t published any titles this year, management has indicated that a design team is still working on the next installment.
I suspect that the loss of this genre has been doubly negative to ATVI. The drag on revenue growth and earnings was the first hit, but also the realization that profitable franchises can disappear has spooked investors. But CoD and Warcraft don’t share much in common with Guitar Hero, in particular the associated hardware requirements and the music licensing fees.
Management is excited about a new Spyro game called “Skylanders”, which uses toys to store game data and access in-game characters. Parents buy a game that comes with a few characters, and then kids beg their parents to buy the rest of the set. Obviously without any real data to go on, it’s difficult to build this into a model, but at the very least it does look innovative. It seems like something I would have liked if I was a kid. More importantly, it’s been totally discounted by the market and at this point represents little more than a free call option.
I believe that part of the reason Skylanders has been so discounted is that it doesn’t target ATVI’s typical audience, and Skylanders’ audience is perhaps assumed to be smaller or less willing to spend on games. Looking at the list of top selling games of all-time, this doesn’t make sense. Nintendogs, a pet simulation game, has sold more copies than Black Ops on Xbox and PS3 combined. The Skylanders target market is viable.
The most important parts of my model have been described above. To value ATVI I’ve built a scenario matrix. I assume that Warcraft can either be flat, down 7%, or down 15% per year. I assume that Activision (CoD) will either be flat, up 10%, or up 20%. For reference, 20% growth is conservative compared to ATVI’s expected console growth rate. Growing sales at the installed base growth rate should be a baseline target for management:
I assume excess cash is used to repurchase shares. Most income statement and balance sheet items grow as a function of revenue. I value the company in 2015 and discount that valuation back to 2012 at 15% annually. My ending earnings multiple is a function of 2015 growth rates, ranging from 6x in the worst case scenario to 14.5x in the best case scenario. The results look like this:
Compared to LTM EPS of $0.84, the valuations above are based off 2015 EPS numbers that look like this:
The matrix indicates that ATVI is probably close to fairly valued assuming modest growth for Activision. But what’s most surprising is the range of values my model produces. This fact coupled with the low price of options means that calls could be an attractive alternative to shares. It’s also surprising to see how much more sensitive the business is to the CoD franchise than to Warcraft. Warcraft can move the needle a few dollars in each direction, Activision can halve or double the stock. Despite Warcraft’s importance to the bottom line today, Blizzard has a clear pipeline of titles and expansion packs to hold investors (and gamers) over until Titan. Also, because we’re looking 5 years out, the compounding growth of Activision easily outweighs Warcraft. My back of the envelope calculations indicated that Warcraft revenue declines would be traumatic, so I’m glad to see that ATVI can weather the storm with respectable growth in the CoD and other Activision titles.
If I decrease my discount rate to 10% rather than the standard 15%, my valuation obviously increases dramatically, not least because this results in a higher ending multiple ranging from 6x to 30x. (Demanding a lower return on investment means I’m willing to pay more for growth):
This set of assumptions is probably more relevant if we’re trying to guess future share prices, and it shows that ATVI is undervalued even with modest CoD sales and major declines in Warcraft.
My model also leaves out numerous key additional sources of upside, including:
- Blizzard movies
- Free to play Call of Duty
- Bungie licensing deal
- Working capital reductions fueling additional share repurchases
- Margin growth from the transition to digital from box distribution
The real play here, in my opinion, is on the option side rather than the stock side. The January 2012 calls at $12.50 cost $0.45 for an Implied volatility of 35%. That seems cheap given that my model conservatively estimates shares are worth north of $20 if Activision can sustain 20% growth despite 15% declines in Warcraft. With the installed base of consoles up 20% and pre-orders of CoD are up over 100% YoY, that target shouldn’t be too hard to reach. January 2013 calls at $12.50 cost $1.15. Both of these expirations have substantial open interest, especially compared to other strike prices and expirations, indicating that I’m not the only person thinking this could be an interesting trade.
Part of the reason the implied volatility is so low is that Actvision has traded within a narrow range for a few years. I believe this is due in part to declining sales from Guitar Hero and the Tony Hawk skateboarding franchises, of which both have been fully worked out with no remaining liability. Nothing else filled that void, and so the real risks of decline come from overestimating CoD or underestimating declines in Warcraft. This chart from Bloomberg compares IV to the share price:
At historically low implied volatility with enormous fundamental reasons to expect volatility in the future, options are a better play than shares. To be clear on how cheap these options are, I’ll walk through an example. Buying enough January 2012 $12.50 options to exercise a portfolio into a 10% position would only cost 0.4% of the portfolio. If the stock closes at $15, the entire portfolio would be up 2.25%. The asymmetry is huge, and even a nominal position can have a significant impact. A $15 guess compares to a $16 price target set by Citi on September 6th, 2011.
Disclosure: Long ATVI