GameStop (GME): optically cheap special situation, but passing

Situation overview

GME is the leading specialty video game retailer in the world. Most are aware that the company has endured tremendous secular pressures over the last several years. It’s been common for 40%+ of the float to be sold short at any given time, yet the company continued to produce solid levels of FCF for many years. With that said, it appears that industry pressures have finally caught up with GME – the stock is down ~75% over the last 2 years.

I’ve followed the company at a high level for some time, but decided to refresh my work given the recent dislocation in the stock and fact that it’s a special situation today. GME actually checks many special situation boxes: (1) a “broken” or failed deal stock, (2) a controversial business repositioning, (3) a tender offer and large ongoing buyback program (authorized to repurchase ~50% of the float at the current price), and (4) a complete dividend cut (leading to the recent dislocation). The story is further complicated by increasing “Blockbuster risk,” recent management turnover, and cyclical pressures and uncertainties ahead of a new console cycle. I grew up an avid gamer, so believe that I understand certain aspects of this business better than most.

I unfortunately think it’s more likely than not that GME doesn’t exist at some point in the reasonable future, but the stock could still be an attractive investment today. The pager industry is a good example of a business model that faced clear technological obsolescence but still delivered attractive shareholder returns for many years. After mergers consolidated the industry and rationalized competition in the early 2000s, the remaining leaders maximized profitability, free cash flow production, and capital returns to shareholders. The businesses focused on sticky government and health care workers and the revenue decline rates ended up being better than feared. Many investors are uncomfortable underwriting these situations, so there exists the potential to purchase industry leaders at extremely low multiples of distributable FCF.

There are many realistic scenarios over the next 3 years where GME returns FCF to shareholders well in excess of the current stock price, but I’m personally struggling to underwrite these scenarios with any degree of confidence. The problem is around SG&A – the business has demonstrated a lack of ability to reduce SG&A in a declining sales environment, and negative operating leverage remains a real issue. For GME to generate the levels of FCF necessary for the stock to work, management needs to cut SG&A at a pace and magnitude that is well in excess of current plans. GME did announce a formal SG&A reduction initiative in Q1 2019, but I find management’s sense of urgency far too low.

Although I’m passing right now, the situations remains fluid and interesting, so I wanted to at least post my initial notes and thoughts. I’m keeping a close eye and my thinking is still evolving in certain areas. If anyone has done work here and thinks that I’m missing anything, please let me know!

The business model & recent events

GME currently runs ~5,800 retail stores globally, with 66% in the U.S., 21% in Europe, 8% in Australia/New Zealand, and 5% in Canada. The key product categories are new physical video game software, pre-owned physical video game software, video game accessories (controllers and headsets), collectibles (pop-culture themed toys, apparel, consumer electronics, home products, and accessories), digital (prepaid currency and codes for online stores), and video game hardware (consoles). Similar to movie theaters, the business is sensitive to the ebbs and flows of AAA game releases, as 20% of SKUs drive 80% of the business. The physical retail presence is the heart of the business model, although GME does operate a website and 60% of store customers first visit the site. The average lease term is just under 2 years, so there’s flexibility on the real estate footprint. Nearly all stores are FCF positive, although management have recently closed ~2% of underperforming stores per year.

Key video game product trends and margins are below (excluding the “Technology Brands” segment):

The average store is 1,700 square feet and produced ~$1.4mm of revenue, ~$390k of gross profit, ~$74k of EBITDA, and ~$56k of EBIT in 2018. When normalizing for working capital swings over the course of the year, the typical store has ~$130k of average invested capital per store, so the business efficiently produces ~30% after-tax returns on capital. Stores generate ~$230 gross profit per square foot, a level of productivity that is above most physical retailers and only modestly below Best Buy, a well managed electronics peer. If not for various technological and competitive threats, GME would comp relatively well against its peer set on key metrics. EBITDA margins were only 5.3% in 2018, however this metric is distorted by the low margin console product and ignores the price guarantees and protections that developers and console manufacturers offer on many new products (somewhat mitigating margin risk).

As the above table shows, GME’s largest profit contributor is pre-owned video game products. To demonstrate how this works, we can look to Gamestop’s website for an example (all figures are as of 6/24/19). Super Smash Bros. Ultimate is a highly popular game released exclusively on the Nintendo Switch console. We can see that the physical game is sold new for $59.99 (the “MSRP” price) or used for $54.99. GME quality-checks and refurbishes trade-ins, giving the consumer comfort against scratches and defects. We can also see that Gamestop is willing to give loyalty members $27.50 of store credit for trading in the game, or $22.00 of cash (non-loyalty members receive less). The reason this model works is because customers are simultaneously willing to (1) sell existing games at a relatively large discount to MSRP and (2) buy used games at a relatively small discount to MSRP. In this example, the spread between the used price and trade-in value is $27.49 – this healthy spread represents GME’s profit potential before processing costs. Different games provide different margin profiles for GME, however the average trade-in economics contribute to very solid gross margins >40%.

From the customer’s perspective, this can also be highly compelling. Many gamers purchase a game, enjoy it for a while, but later move on to the next title. If a gamer purchased this product at $59.99 and later resold it for $27.50, the “effective purchase price” was only $32.49. Only physical games can be traded in to realize this “subsidy,” providing an enduring value proposition to purchasing physical disks over digital downloads. And many gamers are perfectly fine purchasing a used game to save $5.00 off the MSRP – once loaded up, there’s no discernible difference in quality. GME’s focus is optimizing the right used and trade-in prices – both need to be right to simultaneously balance supply (inventory levels) with demand (purchase volumes) and maximize margin (spread x volume). This business naturally benefits from scale and classic two-sided marketplace network effects. As the undisputed leader in secondary video game sales, GME’s has a strong competitive advantage.

Both GME and the customer can win big here, however there needs to be a loser somewhere. In this case, it’s the software developers and console makers that entirely miss out on secondary marketplace economics. It’s worth clarifying that when a PS4/Xbox/Switch game is sold, Sony/Microsoft/Nintendo also receive a cut from the developers. In 2018, this amount of “missed” revenue was $2.5bn at GME, so this gets everyone’s attention. GME argues that their “subsidy” for budget gamers is more additive to demand than anything else, but GME is clearly somewhat profiting at the expense of developers and console makers. The tension between these parties has had varying highs and lows. There’s generally an industry-wide benefit when a specialty retailer exists – especially if the current product trends require consumer education and up-selling potential exists – however Sony/Microsoft/Nintendo need to weigh this benefit against missed secondary economics.

Management hasn’t been blind to the pressures facing the business, and previously tried to mitigate this risk through intelligent diversification. From 2013 through 2016, management aggressively redeployed FCF into building its “Technology Brands” segment, which predominantly consisted of branded AT&T stores. At its peak, this division had 1,500+ stores and did >$800mm revenue. Although this somewhat random growth strategy smelled like “diworsification,” this business did leverage many of GME’s core retail competencies – landlord and real estate management, hiring & training, loyalty programs, and managing a trade-in program. GME often acquired AT&T stores from smaller independent operators that mismanaged the stores. Simple blocking & tackling resulted in immediate profit improvements, and GME generally acquired stores at attractive multiples of run-rate free cash flow (~5x). In addition to realizing retail margins on phone sales, GME collected subscriber management fees (“SMFs”) on the store’s cumulative active subscriber base, providing a high margin recurring revenue stream. AT&T also directly owned and operated ~2,000 stores, creating an aligned-incentive to make the retail distribution model work. AT&T gladly approved GME acquisitions given that GME-owned stores generally exhibited superior volume performance (good for AT&T). For many years, this seemed like a compelling allocation of capital and effective method to de-emphasize physical gaming.

This happy narrative changed in late 2017 when AT&T aggressively changed its retail compensation structure in an effort to prioritize selling and bundling its struggling DirectTV product in retail locations. These changes immediately damaged retail profitability and permanently impaired the segment’s value. AT&T’s decisions seemed highly counterproductive to its own interests, but AT&T management didn’t budge.

To make things worse, there has been significant GME turnover in recent years. GME was led by Paul Raines starting in 2010 – he has a tremendous reputation in the broader retail industry as both a shrewd operator and great capital allocator. GME endured many headwinds on his watch, and it’s clear that the company would be in worse shape if not for his leadership. After stepping down in 2017 for medical reasons, Paul sadly passed away in 2018. The company internally promoted Michael Mauler, however Mauler surprisingly stepped down after only 3 months on the job. Soon after, the Board logically announced that it was running a strategic process.

In the middle of that strategic review, GME decided to sell its Technology Brands division in late 2018. GME sold its stores to Prime Communications (the #2 AT&T reseller) for $700mm. All things considered, the price was decent, GME now had excess liquidity, and management could refocus all of its attention on video games. I also suspected that this deal was strategically done to help facilitate an opportunistic bid for the remaining video game business. This seemed like exactly the type of unloved situation that Sycamore or Apollo would look at.

It turns out that I was wrong (I was wrong many times on this one), and in January 2019 GME announced that they concluded the strategic review without a full company sale. The press release highlighted difficulties in securing acquirer financing at acceptable terms. Given the multitude of sponsor-backed retail LBOs that have recently ended with aggressive covenant abuse (at the expense of creditors), I can’t necessarily blame lenders for balking at this one. George Sherman was quickly announced as the permanent CEO replacement in March of 2019. He brings experience from various well regarded retailers including Advance Auto Parts, Best Buy,┬áTarget, and Home Depot.

Shareholders were naturally frustrated, and management entered into a shareholder agreement for two new Board appointees. GME announced weak Q1 earnings in June 2019, but only made things worse when it simultaneously announced that GME would completely cut the $1.52 annual dividend. The company messaged that it would prioritize prudent de-levering and opportunistic buybacks. The company also issued weak full-year guidance which didn’t help. Although the dividend cut was somewhat expected (the stock was yielding 20% on the existing dividend), the complete removal of dividend support nonetheless hammered the shares and the stock was down 28% that day. The company quickly reacted days later and launched a tender for 12% of the shares. Although this is a large reduction in float, the ~$70mm cash outlay is modest relative to GME’s cash balance and total buyback authorization of $300mm.

It’s been pretty crazy for GME recently! All of these dramatic events contribute to shareholder churn, capitulation, concerns, and potential opportunities. After starting 2017 at ~$25/share, GME is currently ~$5.50/share.

The multitude of secular & cyclical pressures

Now that we’re up to speed on the business, we can spend a bit more time going through the secular and cyclical pressures that GME is facing. We’ll later see that the current price is potentially creating GME at an attractive multiple of near-term FCF. We still need to better understand the duration and sustainability of that FCF, however, so need to do some work on these headwinds.

Shift to digital full-game downloads. Like many forms of media (music, movies, books), gamers have increasingly embraced digitally downloading games instead of buying physical discs. All major consoles have robust online marketplaces where gamers can access all available games. For many years, slow download speeds meant that digital delivery was cumbersome and frustrating, however new pre-loading technology allows for gamers to instantly enjoy a newly released game at midnight. GME does participate by selling digital currency and download codes in its stores, however its market share is much lower than traditional physical discs.

There is a lot of conflicting industry data on this topic, but most sources suggest that that the industry went from ~80/20 physical/digital in 2009 to ~20/80 today. This data is a bit misleading, because it’s obfuscated by PC and mobile gaming (which are entirely digital), microtransactions, and subscription models (more on this later). GME’s core focus is AAA console titles ($59.99 MSRP), where the digital penetration has slowed and stagnated at much lower penetration rates. Different sources will tell you different estimates, but I think AAA console titles are only ~40-50% digital today.

The decline of console developers. This trend is a bit more anecdotal, but I still believe it to be a key headwind. In the industry’s early years, the barriers-to-entry in making a game were relatively low and there was large diversity in developers and games. Doom, an extremely popular shooter released in the early 90s, was developed by a team of 7 friends. Over time, however, the increasing technological horsepower of consoles has significantly raised costs for developers. Better chips have improved graphical fidelity and most publishers have responded by throwing increasing numbers of artists at the challenge. Development costs began to grow exponentially. For perspective, consider Red Dead Redemption 2, a top AAA game that was recently released from Rockstar Games (the Grand Theft Auto developer that is owned by Take Two). Red Dead contains an extremely character-rich and immersive single-player experience that tells a story about outlaws in the old west. The game was released to exceptional critical praise, however it took 7 years to develop, required a peak team of 1,000 developers, contains 300,000 character animations, and includes 500,000 lines of dialogue produced by 1,200 voice actors! This came at tremendous cost (in the hundreds of millions of dollars), but Take Two could justify these costs because the game sold $725mm in its first 3 days alone.

“Mid-tier” console developers have been increasingly unable to profitably absorb these costs and have either exited the industry or were acquired by “Tier 1” developers. This has pushed many simpler indie games away from consoles to mobile, a headwind for the console-focused GME. The rising budgets for AAA console games have also had other impacts. Most notably, higher budgets have increased the risk of development and penalty for producing a flop, and most developers have de-risked production by focusing on sequels in established franchises. Some have compared this to the movie industry. Initially, the movie industry was an area for creativity and artistic expression. This changed in the 70s with Jaws and Star Wars, as movie producers saw the “summer blockbuster” as an increasingly risk averse way to profitably deliver a movie with high production quality.

Increased digital DLC and microtransaction models. The increasing development lead times of many AAA games initially led to boom-bust profit dynamics over a release cycle, so many developers logically tried to smooth-out profits and introduce recurring revenue sources. This first came from downloadable content (“DLC”), where additional levels and expansions were sold post-release at price points of ~$10-$30. Next came microtransactions, where smaller character cosmetics and items could be acquired for ~$1-$5. On the PC, many successful games were entirely subscription-based. More recently, many popular PC games are even free-to-play (“F2P”), where developers make all of their money on optional microtransactions. These games are generally arena-based multiplayer games – costing much less to make than AAA games – making the economics work.

Console developers first tried to have their cake and eat it too – they added DLC and microtransactions, but also maintained the up-front $60.00 MSRP. Video games generally offer great value when compared to other entertainment sources (on an hourly basis), but console gamers still became increasingly frustrated when looking at F2P games on PC. EA’s aggressive use of microtransactions in Star Wars Battlefront 2 resulted in widespread boycotts and consumer backlash (EA has since moderated its level of microtransactions for $60.00 games). Making matters worse, we’ve also seen the rise of the highly popular “battle royale” games like Fortnite. Fortnite was the first blockbuster game to bring the F2P model to consoles – gamers play Fortnite for free, and the developer makes all of its money from entirely optional cosmetic character outfits and dances. Fortnite was a relatively low budget game to develop, so the F2P model certainly works better than in other instances. By comparison, Red Dead Redemption 2 couldn’t realistically be F2P – the up-front costs are too high, and loyal customers are fine paying $60.00. With that said, continued consumer conditioning and reception to the F2P model will only further pressure other developers to consider this route.

All of this is bad for GME because GME’s market share of up-front $60.00 game sales is much higher than its market share of DLC, microtransactions, and F2P revenue. Remember, many AAA console games are still sold as physical discs, and GME is the leader in this category with high market share. Once that physical disc is sold, however, GME is much less successful at fully capturing that consumer’s DLC and microtransaction spend. GME will try to bundle and attach digital currency sales to its in-store transactions, but most consumers inevitably hold-off and later make these digital purchases in-game.

Software developers have welcomed many of these trends, encouraging long-tailed engagement on releases and the associated long-tailed recurring revenue contributions. One of Activision’s biggest esports bet – the popular game Overwatch – still receives consistent updates, new levels, and competitive support despite being released in 2016. Wall Street loves recurring revenue, however the declining emphasis of up-front $60.00 MSRPs has only pressured GME’s business model.

Increased retail competition. This is also anecdotal, but I believe that Amazon has recently increased its focus on physical and digital gaming. I’ve even seen Amazon-exclusive discounts for certain titles.

Late-cycle pressures and tough comps. It seems like the next major wave of consoles from Sony/Microsoft will come in 2020. Whenever new systems are on the horizon, gamers generally opt to hold off on console purchases and wait out for the new systems. This can make the final 12-24 months of an old console cycle a bit painful for retailers like GME. Hardware sales are much lower margin than other key products, but GME is still feeling some pressures.

Threat of new consoles (including cloud). New consoles do bring some degree of opportunity and excitement for GME, but there’s also an inevitable risk that a step-change in technology will de-emphasize or even entirely eliminate physical games. The business model for Sony/Microsoft/Nintendo is to develop compelling hardware, sell this as cheaply as possible to drive and grow an install-base, support game developers, and earn a small but high margin cut of every game sale. It’s the razor-razorblade business model.

And unlike physical discs, digital games cost nothing to “manufacture.” These sales are facilitated by digital storefronts operated by Sony/Microsoft/Nintendo, and developers share higher economics with console makers as a result. This means that both console makers and developers should strongly prefer further digital share gains: (1) both groups receive higher profits per game, and (2) both know that digital sales cannot be later sold in the secondary market. It’s therefore reasonable to assume that console makers and developers will encourage further digital share gains. Microsoft has historically been more vocal about this goal.

There were technological barriers preventing this for many years, although restraints like internet speeds have been weakened. Xbox even tried to accelerate the digital shift during its prior Xbox One console release in 2013. Microsoft initially planned to require a console to frequently “ping” Microsoft servers for the console to run a particular game. The benefit of this “always online” model is that it would allow Microsoft to verify game ownership, hampering secondary trading. Sony didn’t follow suit, and extreme consumer backlash resulted in Microsoft walking back these plans. Many gamers never fully forgave Microsoft, and this was a contributing factor to Sony’s initial momentum that led it to easily outsell Xbox over the current console cycle.

So maybe console makers have learned and will continue to fully support physical discs? Perhaps, although Microsoft recently released a “digital-only” version of its Xbox One. The lack of a disc-reader reduces size and cost. It’s worth remembering that ~100% of PC games are downloaded digitally. There’s nothing really stopping consoles from going this path, although any transition needs to be gentle.

Going one step further, Google has recently announced its fully cloud-based Stadia console. The idea is that the current system of video game hardware – where consoles are decentralized in people’s homes and updated every 5-10 years – is inefficient. Like many resource-intensive computing applications, gaming could be more efficient if the computing power is centralized and scaled on Google’s cloud servers. Gamers will still have their controller, but button presses will be transmitted to Google’s server and sent back to your TV. None of this is possible without recent improvements in internet speeds. Over time, Stadia performance and graphics will improve with chip advancements. To address the Stadia threat, Microsoft and Sony have made the unprecedented decision to team-up on their own Azure-based cloud partnership. It’s unclear the ultimate goal here, but Microsoft and Sony clearly see cloud gaming as a real opportunity and Google as a threat. Sony already has a cloud-enabled subscription-based streaming service for older games (not new releases), although the quality of that experience has received mixed reviews.

Although many forms of media have transitioned to streaming formats – including TV, music, and movies – there are constraints to this happening overnight for gaming. The biggest issue is latency – the delay between a gamer pressing a button and their in-game character performing the action. In a traditional console experience, latency is virtually nonexistent; in a cloud-enabled experience, latency becomes a problem. Google and others promise that users with high-speed internet connections will receive minimal latency, but gamers have a way of sensing any material delays. For online-based multiplayer shooters (where local PCs/consoles are communicating with servers), a relative difference of 25-50 milliseconds is considered a recognizable disadvantage. Initial reviews for Google’s Stadia experience were somewhat mixed, and many criticisms were focused on the “feeling” of it. Cloud gaming remains a promising and intriguing future for gaming, but I don’t think we’re there yet.

Some mitigating factors

There are some characteristics of GME’s business that help mitigate these headwinds. I’ll walk through a few below.

Natural constraints in digital penetration. Although console makers and developers want to encourage high margin digital adoption (through discounting and new revenue structures), there are some natural constraints in the player base and they need to be careful about alienating console players (like Microsoft did in 2013). First, it’s helpful to look at other media forms as “case studies” for ultimate digital adoption rates. Books have settled at ~30% digital penetration, music at ~50%, and movies at ~50%. And unlike AAA games, these medias are smaller file sizes and contain immaterial residual value. It’s therefore reasonable to assume that AAA games have a natural digital adoption rate that is close to 50%, a level that is relatively consistent with the current adoption rate. Put differently, despite many of the technological barriers being lifted on digital game downloads (pre-downloading, internet speeds in most areas, etc.), digital penetration for AAA games only climbed so high. To go much higher, it probably requires console makers to “force” adoption. Forced adoption remains a real but different risk.

Many have polled gamers as to why they still buy physical games, and the results are interesting. Most actually say that they simply enjoy buying the physical discs because they like having the collection of physical boxes on their shelves (similar to book or album collectors). Gamers also frequently cite the “subsidy” that can come from later reselling their games in the secondary market. Gamers can also bring a physical disc to their friend’s house to try out the new game. A common complaint is the size of digital games. Digital proponents (and GME bears) often cite the significant increase in console hard drive sizes, but this ignores the fact that much of the hard drive is consumed by non-game files and recent games have similarly increased in size. For example, the Xbox hard drive is 500 GB, but only 400GB is available for games. This sounds like a lot, but Red Dead Redemption 2 alone requires 107 GB! Not every game is that big, but hardcore gamers can quickly run into storage constraints. The Nintendo Switch is less technologically demanding, but nevertheless has a small base hard drive at 32 GB. There are already individual Switch games that exceed this size.

Remember that console makers are trying to sell the razors as cheaply as possible to maximize profits on the razorblade. They will always be tempted to skimp on the hard drive space and keep up-front console costs low. I think digital penetration is much lower on the Nintendo Switch, and this console’s continued success and install base growth should provide a tailwind to GME.

PowerUp loyalty program. GME has an expansive and well-designed loyalty program, and I think this has been critically important to GME. This loyalty program is called PowerUp in the U.S., and other similar loyalty programs exist in international countries. PowerUp currently has 39.6mm total registered members, 16.2mm active members (shopping at GME within the last year) and 5.6mm paying members. Outside the U.S., GME has another 20.9mm registered loyalty members. PowerUp members are generally hardcore gamers and high value customers. PowerUp members produce 3x the revenue of non-member customers and 5x the profits. They frequently trade-in games and drive that two-sided secondary ecosystem.

Non-paying members receive points for every dollar spent, but no material additional benefits. Paying members ($14.99/year) receive 2x the points for every dollar spent, get 10% off pre-owned games, 10% extra trade-credit when selling games and hardware, an annual subscription to Game Informer magazine, and additional promotions. For active gamers that frequently buy and sell used games, the benefits can quickly exceed $14.99/year. It’s no surprise that GME focuses its rewards on encouraging the pre-owned ecosystem, as pre-owned margins are its highest product.

Paid loyalty programs can be powerful when executed well, and the greatest success story is probably Amazon Prime. For a high volume online shopper, the annual Prime membership fee is easily exceeded by the value it brings. And once that customer pays for Prime, they proactively try prioritize shopping on Amazon as much as possible. PowerUp is similar in that it encourages GME’s high volume customers to do all of their purchases through GME. These gamers are also more likely to continue purchasing physical games due to the enhanced trade-in economics. And if they truly want to transition to digital products, they are more likely to purchase digital currency and codes through GME.

Most importantly, this loyal customer base ultimately drives 70% of GME’s revenue. This means that the vast majority of GME’s revenue is driven by customers that are incentivized to continue prioritizing GME. This truly helps mitigate some of the secular pressures and helps extend the melting ice cube a bit. This all sounds great, however my main concern is that paying members declined by 10% in 2018 – this will be a key metric to track.

Thinking about value

A key question remains: does this business need to exist? I don’t think the gaming industry necessarily needs a specialty retailer, so I think the answer is “no.” Gamestop does add some degree of value to the industry (its employees are all gamers and can educate and up-sell customers), but I also see no reason why consoles/games couldn’t be exclusively sold through Amazon/Walmart/Best Buy. Making things worse, games probably transition to fully digital at some point (in a gentle way that minimizes frustration in the customer base), and most other key parties (console makers & developers) would prefer to see this. PC gamers get by just fine in an entirely digital environment. GME cannot be profitable on hardware and accessory sales alone, so I struggle to see any realistic scenario where this is a going concern 10 years from now. I think investors need to think about value in a similar way.

2019 is a bit of a unique year given the planned de-levering, large capital return program, and relatively pronounced cyclical pressures ahead of next year’s console launches. For these reasons, I’ll frame valuation from the perspective of year-end 2019. A large piece of the excess cash will be deployed at that time, and we will potentially be looking at conservative LTM numbers from a cyclical perspective.

A reasonable first question is: what’s the liquidation value of this business? In an “orderly” and gradual liquidation, GME’s “net-net” balance sheet value would be ~$3.50/share at Q4 2019. This generously values inventory and receivables at face, conservatively values GME’s 1.1mm square feet of owned distribution facilities at $0, makes no adjustments for rejected leases (which appear in-line with market rents at ~$35 per square foot), and assumes no bankruptcy rejection of outstanding customer store credit. This is a highly simplified analysis, but clearly suggests value below the current stock price. Anything less “orderly” – a.k.a. an actual liquidation – would only result in lower recoveries to equity holders. There are liquidation sites online where you can see discounted video game values, and it’s not great. So despite GME’s dramatic stock sell-off, the $5.50 share price is still not yet supported by balance sheet value.

GME is still generating free cash flow, however, and isn’t liquidating tomorrow. This means that the business should continue to generate a declining stream of free cash flow, return that cash to shareholders, and potentially provide value to shareholders in excess of its balance sheet value. This is the “melting ice cube” line of thinking. This analysis is somewhat complicated by the seasonality of working capital, remaining excess liquidity following the Tech Brands divestiture, and current capital return program.

Let’s first talk about revenue. Video game revenues have been ~$8.5bn over the last 5 years, however revenues were -3% in 2018 and 2019 is looking even worse. The company has guided to -5-10% revenue declines for FY 2019, but I think this is actually too optimistic. The prospects of new consoles have frozen hardware sales (-35% in Q1), this year’s AAA release lineup comps very weak vs. last years holiday releases, pre-owned game volumes are continuing to follow new unit sales lower, accessories face a difficult comp vs. last year (the rise of Fortnite’s popularity produced strong headset sales in 2018), digital sales are inflecting lower (potentially on lower PowerUp membership), and the rare bright spot – collectibles growth – has slowed to +10%. Taken together, I model revenue -15% for FY 2019, worse than the lower end of guidance. On the positive side, there should be some modest mix benefits to consolidated gross margins.

Against this increasingly challenging revenue environment, I see continued operating leverage pressures from SG&A. When GME was a $9.1bn revenue business in 2014, there was $1.9bn of SG&A (72% of gross profit). Revenues had declined to $8.3bn by 2018, yet SG&A still remains at $1.9bn (81% of gross profit). The new CEO has pledged a $100mm SG&A reduction program over 2019 (a mere 5% cut), but this still implies negative operating leverage against any realistic revenue declines. $100mm is simply far too low of a target. After years of stagnating SG&A levels and accelerating business model threats, the level of urgency needs to be much higher. Assuming another ~100-125 store closures and the gradual realization of the $100mm reduction initiative over the course of the year, I model 2019 SG&A being down $90mm year-over-year.

The company ended 2018 with $1.6bn in unrestricted cash, which includes the Tech Brands proceeds. It’s worth noting that Q4 is a seasonally low working capital quarter, so this balance overstates the true excess cash levels. After paying off the $350mm 2019 notes in Q1, I model the company buying back another $250mm of the 2021 notes ($475mm issue) to achieve its 1.0x gross leverage target and further reduce interest expense. The company’s current tender offer implies a ~$70mm cash outlay, but this still leaves a lot left on the $300mm buyback authorization. I’ll assume that additional buybacks brings the total to $100mm for the year, implying ~18mm shares repurchased (~18% of the float). The actual magnitude and allocation of debt reduction vs. buybacks could certainly turn out different.

Taking everything together, I model 2019E adjusted EPS of ~$0.90, well below street at ~$1.60. To the extent revenue “only” declines -7.5% (midpoint of guidance), then EPS could be as high as ~$2.00. The extremely wide gap is illustrative of the high sensitivity to both positive and negative operating leverage. Without a well thought out, high conviction, and tangible plan to reduce SG&A much further, it’s impossible to have any degree of confidence in GME’s earnings power. For perspective, GME was doing ~$3.35-$3.50 of earnings before divesting Tech Brands. After that divestiture, GME did $2.79 in 2017 EPS and $2.14 in 2018 EPS. Although industry pressures have existed for some time, the financial pressures have accelerated recently.

Let’s say that my pessimistic view is right and GME indeed does ~$0.90 of 2019 EPS. GME’s free cash flow conversion is relatively clean, so we can assume that FCF is also $0.90/share or ~$85mm. After my $600mm modeled debt reduction, $100mm buyback, and $85mm FCF generation, GME would have $1bn of cash at Q4 2019. If we look at seasonal working capital needs and adjust Q4’s abnormally low working capital level for average working capital needs, we see that this $1bn projected cash balance is really ~$625mm when seasonally-adjusted. And from that $625mm seasonally-adjusted cash balance, I think the company would prefer to keep at least $200mm of that on the balance sheet – this represents the additional cash needed to build to peak working capital needs. So we probably have ~$425mm of gross excess cash or $200mm net excess cash after deducting the remaining $225mm 2021 notes. This means that GME could exit this year with ~$2.40/share of net distributable cash that it could pay shareholders.

At a current stock price of $5.50, we can back out this $2.40 of projected net distributable cash and create the underlying unlevered enterprise at $3.10/share. Put differently, at the end of this year we could be creating the $0.90 of FCF at a ~30% LTM FCF yield.

Is this attractive? It all depends on our view of the $0.90 in FCF. In some ways, the $0.90 estimate is conservative – we are only giving partial-year credit for my projected reductions in interest expense and shares outstanding. The full-year annualized benefits should provide an incremental ~$0.50 tailwind to 2020 earnings. In addition, we know that 2019 is a cyclical trough ahead of the new 2020 console cycle. The 2020 AAA release lineup also looks much better vs. 2019’s weaker lineup. Taken together, it’s not unrealistic for 2020 revenues to remain flat or even modestly grow – this would provide a much needed breather against operating leverage pressures. I’m also optimistic that management will become laser-focused on further SG&A cuts and find additional areas of opportunities following the 2019 review. If management is able to again cut 5% of SG&A, this would drive an additional ~$0.85 of EPS. In bullish scenarios, we could therefore see 2020 EPS as high as $2.25 ($0.90 + $0.50 + $0.85).

With all of that said, my confidence in forecasting 2020 remains extremely low, there is a wide range of outcomes, and I think things are biased to the downside. If we look at the prior console launch cycle in 2013, gross profit was actually modestly down – GME failed to participate in the industry’s growth. If revenues again declined at punitive decremental margins, we could see EBIT margins quickly approach 0% (margins were 4.0% in 2018). And although the new console launch cycle brings several opportunities, it also presents increased tail risk that new features accelerate digital adoption. We should see hardware sales do well in 2020, but this is GME’s lowest margin product. GME only makes 8.5% gross margins on the “razors,” so the net benefits are modest without an associated software boost.

Let’s think back to the mid 2000’s pager example. In that case, we had a continued product necessity for government and healthcare jobs, ensuring that the melting ice cube was orderly and gradual. We also had significant industry consolidation, ensuring highly rationale competition in those final years. There was real value to that slowly declining FCF stream and you could value it with reasonable accuracy. GME is sadly not like that. We have high uncertainty around the continued necessity of the business, no view around the duration of the ice cube, and tremendous competitive threats and uncertainties. This would be entirely different if management boldly and confidently said: “yes, over time we could eventually cut 25% of SG&A if needed.” Without that promise, we can’t handicap the operating leverage risk.

Making matters worse, I disagree with the elimination of the dividend. If I was management, I would immediately pay off all of the debt – Toys ‘R’ Us showed the tail risk if suppliers become concerned about a retailer’s balance sheet. After that, I would distribute all remaining excess liquidity through a special dividend (our $2.40/share estimate), and promise to pay out 100% of future FCF production in the form of a variable dividend. After years of buying back stock >$20.00, management needs to show that the melting ice cube will actually be distributed to shareholders. Although there is some degree of uncertainty in a variable dividend, the value proposition is significantly more certain vs. buying back stock at $5.50.

In many ways I’m rooting for GME – I was a customer growing up. And it’s not crazy to see the stock doing well. I laid out a bullish case where investors could get $4.65 in dividends through 2020 ($2.40 special dividend + $2.25 2020 FCF), or 85% of the current stock price. From there, investors would have close to a free look at GME continuing to generate cash in the early years of the next console cycle. The stock can certainly work, but I’m not convinced. My opinion is that the potential upside of this situation fails to properly compensate you for the real and likely risk of capital impairment. To the extent the stock ever approached liquidation value, the risk-reward profile would be notably more compelling. We have a ways to go.

This situation remains extremely interesting for a variety of reasons, but I’m simply watching for now.