Description

PLAY: Customer traffic recovery at only 40% of pre-COVID trend at stores reopened for at least a month according to transaction, geolocation, and Google Maps data. 60% base case downside. Further equity raises are likely needed this year to avoid covenant breach.

Company overview

Dave & Buster’s (D&B; ticker PLAY) is a big-box (~41K Avg. SqFt.) arcade and casual dining chain with 137 locations across 39 states nationwide. ~60% of revenue comes from the Amusements segment (arcade), while ~40% comes from the on-premise Food & Beverage segment.

Thesis

PLAY’s recent rally has priced in an optimistic productivity recovery scenario and brought dilution-adjusted valuation recovery to in-line with casual dining comps. However, we believe PLAY’s productivity will remain impaired for the foreseeable future in both absolute and relative terms because it is one of the worst positioned consumer concepts for a post-COVID world. Analysis of PLAY’s revenue and customer traffic trends supports our view that PLAY’s productivity recovery will have a slower pace and lower ceiling than current valuation implies. PLAY’s high fixed cost operating model implies severely impaired EBITDA and cashflow, starving the company of capex and putting same store sales in further jeopardy during the out years. We project a base case 2021 EBITDA of $160mm, suggesting a stock price of $7.5 at 6.5x, the average 1.5yr fwd multiple PLAY traded for during 2019.

Research methods

We specialize in the consumer vertical and utilize a combination of alternative data and fundamental primary research methods. See the Appendix for more details on our PLAY research.

Recent stock rally implies a strong rebound in store productivity by 2021…

-          Why PLAY rallied >40% last week

o   Casual dining chains rallied last week, driven by optimism about ongoing economic reopening, speculation about a COVID vaccine, and bullish commentary from Cheesecake Factory (CAKE) regarding their Q2 same store sales.

o   We speculate that PLAY squeezed higher on optimism that the same factors which are driving casual dining performance will allow it to recapture its prior productivity faster than previously expected.

-          Frames to think about what current stock price implies

o   Frame 1: PLAY will recapture ~97% of FY19’s productivity in FY21, without further equity dilution.

o   Frame 2: Either PLAY is no more exposed to COVID anxiety than a typical casual dining chain or that it is virtually certain that COVID anxiety fully recedes over the next 6-12 months.

§  Adjusted for dilution, PLAY is now down -24% YTD, exactly in-line with the average bellwether restaurant comp group of CAKE, EAT, CBRL, DENN, DRI, and BLMN. See Relative Perspective chart in Appendix.

…but store productivity is likely significantly impaired for the foreseeable future

-          D&B’s primary traffic driver (arcade) will remain pressured while COVID anxiety persists

o   Arcade visitors move randomly from game to game, many of which require touching objects/controllers or putting equipment on one’s face (VR). Sanitizing each game between users would require extraordinary measures which would be labor-intensive and/or degrade the user experience.

o   Social distancing would be impractical as many of D&B’s games are cooperative, with no physical way to create distance between players even when limiting the total number of customers allowed in the arcade.

o   39% of D&B’s store fleet is in malls, which suffer more than off-mall shopping centers in the post-COVID environment.

o   Other amusement concepts (movie theaters, theme parks, bowling, pool, TopGolf, go karts, trampoline parks, etc.) have better mitigation options because they can either enforce social distancing or don’t require visitors to interact with objects that someone from outside of their group just touched.

o   See Appendix for images of the D&B consumer experience

-          D&B’s Food & Beverage segment will miss out on the off-premise trend

o   While it is plausible that casual dining could regain lost productivity in 2021, we believe it would include a significant mix shift toward off-premise and delivery. Only a minority of D&B’s restaurant customers go there primarily for the food, so D&B’s off-premise business is practically non-existent.

o   Alix Partners’ COVID-19 consumer survey from late April suggests PLAY is poorly positioned post-quarantine relative to casual dining peers due to the lack of an off-premise offering:

-          Quarantine-induced home gaming/VR hardware spend may accelerate erosion of arcade gaming

o   Quarantine has caused an unprecedented acceleration of spend on in-home gaming and virtual reality equipment (see exhibits below). It is likely that in-home gaming was already contributing to PLAY’s negative same store sales in FY18-FY19. Much as the proliferation of big-screen TVs has eroded movie theater attendance, it would be surprising if the extreme acceleration of in-home gaming/VR spend had no impact on arcades. See exhibits below.

o   According to the CFO of Best Buy, this acceleration in home gaming is more likely to be a permanent step-change than merely a pull-forward of demand:

§  “Gaming is the perfect [example]… that's not pull-forward demand, that demand that would not have existed; people would not have bought those gaming consoles. They would have waited for the next generation of consoles to come in December, but because you're in this unique life situation, there is real incremental demand there.” – Q1 Earnings Call

Users connecting to Steam via Virtual Reality (VR) gear

-          D&B will be starved of capex, putting comps in jeopardy in the out years

o   While PLAY’s ultimate productivity recovery ceiling is uncertain, it is difficult to imagine that PLAY will have cash to invest in significant new game introductions over the next year. We think this will futher impair PLAY's ability to drive traffic through and post-COVID.

o   FY18 – FY19 saw a steady stream of new product introductions, including Virtual Reality, which was the single biggest amusement investment PLAY has ever made and a major traffic driver. See Appendix for FY18-FY19 new game introductions calendar and management quote.

Insights from recent data analysis

Our analysis suggests D&B’s relatively slow re-opening cadence is likely influenced by low consumer demand

-          Not re-opening stores even when possible

o   As of June 5^th, D&B has opened only ~25% of its total stores and is lagging competitors in openings within its most important states. See exhibit below.

o   D&B has deferred reopening many stores in states where they are eligible to reopen. We believe this caution is driven by weak demand in reopened stores, combined with the high fixed cost of store operations (more on this in our modeling assumptions section).

-          Underperforming relevant categories and competitors

o   D&B’s revenue recovery is lagging its most comparable NAICS categories in the South region where it has opened the highest proportion of its stores. D&B’s most relevant category (Amusements) is itself lagging in recovery pace. See exhibit below.

o   D&B’s chainwide visitor traffic recovery (inclusive of closed stores) lags its most oft-cited competitor (TopGolf) even though TopGolf has re-opened a slightly smaller percentage of its store fleet to date. See exhibit below.

Note: South region states include: TX, FL, GA, AL, OK, AR, KY, LA, TN, VA, MS, DE, MD, NC, SC, and WV

-          Visitor traffic cohort analysis for re-opened stores suggests muted demand

o   Our visitor traffic data is reliable for 19 of D&B’s 28 reopened stores. This data suggests reopened stores recover on average only ~40% of their pre-COVID YoY visitor traffic trend one month after reopening, with the recovery slope slowing significantly after the first week. See exhibit below. Also see Google Maps location crowding data from June 5^th in the Appendix.

§  Note: The Panama City location is an interesting outlier, which has recovered ~87% of its 2019 baseline productivity. However, since it was running up ~58% YoY pre-COVID, the recovery versus pre-COVID YoY trend is ~55%.

o   If our data is correct, PLAY’s recovery pace for re-opened stores lags dramatically behind announced casual dining comps from recent earnings calls:

§  Outback Steakhouse’s stores with open dining room have recovered 83% of productivity

§  Cheesecake Factory’s reopened stores have recovered 75% of their productivity

§  Olive Garden stores have recovered 74% of their productivity by May 17^th

§  Cracker Barrel units with open dining rooms recovered 68% of productivity by late May

o   While we expect D&B’s recovery to continue in absolute terms, we believe the current trends support our view that D&B’s recovery is likely to continue to lag far behind the casual dining sector for the foreseeable future.

 Modeling assumptions & valuation

Appreciating Operating Leverage – limited FCF until mid-way 2021

Unsurprisingly, PLAY is an asset heavy business, which comes with its associated amount of fixed costs. With boxes that average 41K square feet, it has one of the highest rent burdens in all of restaurant public markets. Less obvious is the fixed cost of operating the arcade (utilities and maintenance) and the minimum staffing required to operate the restaurant, bar, arcade floor and VR equipment under a lit scenario.

We have done our best to break down the restaurant-level cost buckets into fixed vs variable cost. While we don’t purport precision, we have done our best to be conservative and the below adequately illustrates the point: restaurant operating margins are quite sensitive to our productivity assumption. At the bottom of the table we have illustrated the implied restaurant level margin in each of our 2021 cases. 

As a ballpark rule of thumb, assuming FY19 SG&A (excluding stock-based comp), an Interest burden of ~$28mm and maintenance capex of ~$40mm, productivity (across the year) needs to be ~85-90% of 2019’s to be free cash flow breakeven.

As a sanity check and an indication of how conservative we are being: across FY18+FY19, the 2yr comp stack was -4.2%, which came alongside ~365bps of restaurant-level margin deterioration. All that we’re claiming in our base case is that, if productivity is -15% from FY19 levels on average across FY21 (implied via our ’20E & ‘21E exit-rates of 80% & 90%), restaurant-level margins will be -9% from ’FY9. This implies that the observed ~85bps per point of comp decline in restaurant-level margins in FY18 & FY19 decreases to ~60bps per point of comp decline going forward to reflect substantial store-level cost cuts.

We encourage investors to model operating leverage thoughtfully, and sensitize what run-rate FY20E & FY21E would each have to exit at in order for PLAY to generate cash during FY21.

Liquidity: Equity financing likely required to avoid minimum liquidity covenant in 4Q20

As of 4/30/20, PLAY had ~$263mm of cash on hand, pro forma for all equity issuances in May. Management stated that its weekly burn was $7.2mm including rent, $3.9mm excluding rent, with staff furloughed and the footprint fully dark. Admittedly this analysis is unsophisticated, but it illustrates the point: if we assume weekly burn gradually improves as stores reopen but productivity of re-openings remains below restaurant-level FCF breakevens of ~75%, partially offset by un-furloughed employees, and if rent is to be paid once stores re-open as well as the gradual repayment of rent in arrears, we believe PLAY will trip their minimum liquidity revolver covenant during 4Q20 – before the net leverage covenant gets reinstated for quarterly testing - and ultimately run out of cash in 1Q21. 

We found it particularly interesting that alongside the most recent revolver amendment lending banks introduced this new concept of a minimum liquidity maintenance covenant. We believe this may be a reflection that they felt it necessary to build in incremental protection. We believe PLAY should and will tap the equity markets well ahead of testing this covenant. Further, if our prophecy comes true, we encourage the market to scrutinize the timing of the offering - how far in advance of a potential breach PLAY taps the market, as well as at what share price - as an indication of management’s truest view of the business’ outlook (note: recent equity raises were ~$10/sh vs today’s $20).

In our base case we expect a further $100mm of dilution to equity holders, ~11% of the current market cap to avoid breaching the minimum liquidity covenant.

Interestingly, BJRI explained to us on their earnings call that when they re-opened dining rooms in Texas in stores whose kitchens were already servicing off-premise, at only 25% capacity utilization the burn actually grows. For the reasons detailed above, we believe PLAY’s cost structure sets the bar significantly higher than a restaurant that’s only incrementally opening its dining room. 

BJRI: An analyst asks: “I just want to be clear... those stores saw a greater cash burn after the dining room opened? Is that correct? to which CFO Greg Levin responds: “No. So just -- what we're saying is, don't expect us to convert to a wildly different burn rate immediately based upon that sales increase because of the cost of, really, the incremental cost of opening at those volumes, right? …. the cost of the supplies, et cetera, is not going to wildly change the $2.5 million burn rate until that number grows. That's really all we're trying to say there…. We've realized the cost of the supplies… it's not going to be wildly incremental at those numbers -- may put a little short-term pressure, but it's not -- we wouldn't be doing it if we thought we were going to change our burn rate significantly.”

All in, this gives us comfort that, while productivity remains <75% of 2019’s, PLAY’s burn rate will likely remain elevated. 

Risk/Reward & Valuation Framework:

Our Base/Up/Down cases suggest stock prices of $7.5/$0/$24, respectively, or -63%/-100%/+17% from Monday’s open price  of ~$20/sh.

Assumptions:

·         We use 6.5x our ‘21E numbers to reflect the average of the 6.7x 1yr fwd EV/EBITDA and 6.3x 2yr fwd EV/EBITDA that PLAY traded at during 2019 (data available in appendix).

·         Base case $7.5 (-63%): FY21 achieves 85% of ‘19’s AUV (80% ’20 exit rate / 90% ‘21E exit run-rate); 18% RLM; $100mm equity raised; 6.5x EBITDA.

·         Upside case $0: FY21 achieves 77.5% of ‘19’s AUV (70% ’20 exit rate / 85% ‘21E exit run-rate); $150mm equity raised; 10% RLM; stock isn’t viable (requires >10x EBITDA)

·         Downside case $24 (+17%): FY21 achieves 95% of ‘19’s AUV (90% ’20 exit rate / 100% ‘21E exit run-rate); 24.5% RLM; $50mm equity raised; 6.5x EBITDA.

Valuation Considerations:

Net Debt: Project the balance sheet. The biggest mistake we see in sell-side’s price target framework is failing to pro forma the balance sheet. The Street has Net Debt coming down over the NTM, despite the fact that this Company will clearly burn cash and therefore Net Debt should come up. Every $100mm PLAY that burns costs equity holders ~10% on the stock. With ~$263mm of susceptible cash on the balance sheet and our expectation that much of this will be burned before a stabilization can be achieved, running this forward is significant.

EV/EBITDA: D&A almost half of EBITDA, and may come down due to impairment. We’ve seen both buy- and sell-side default to thinking about PLAY on an EV/EBITDA basis. With no EPS / FCF, we understand that it’s the most practical methodology and therefore use it ourselves in our price target framework. However, in 2019, D&A of $132mm represented ~43% of EBITDA. We believe it will be closer to 90% of EBITDA in 2021, and note the implied discrepancy between EV/EBITDA on 6.5x equates to an EV/EBIT of ~50x. Separately, We believe PLAY may take impairment charges in the coming quarters, which will translate into lower fwd D&A & be a drag on both EBITDA & ultimately enterprise value.

Parting Words: sanity checking our 2021E exit rate

2021’s exit-rate productivity relative to 2019’s AUV conceptually is equal to 1 minus the 2-year comp stack. Therefore, we ask the following question: if we were to completely ignore COVID, but assume that capex is materially depressed over the next 12-24 months, what would PLAY comp across the next 2 years?

Baseline: PLAY comped -1.6% in FY18 and -2.6% in FY19, resulting in a 2yr stack of -4.2%. Across these two years, PLAY had a robust pipeline of introductions, including their first virtual reality offering, which we noted above was the single biggest amusement investment the Company has ever made (see quote in appendix).

As we stated above, based on management commentary we can assume PLAY took ~1.5% pricing in amusements during most of FY19. Extrapolating this to both years (3%), and because amusements represents 60% of the business, we can assume arcade pricing alone contributed ~1.8% to the total 2yr stack comp across FY18 & FY19. Importantly, we note this excludes any potential benefit that the business received from these introductions driving incremental traffic.

So if over the past 2 years the business comped approx. -6% excluding price taken in the arcade, we believe something in the ballpark of our base case – a 2yr stack of -10% which gets us to 90% productivity - is reasonable to believe in the circumstance of depressed capex and lack of traffic drivers. While we could argue for more, we think this is all one needs to believe to appreciate the material downside that we see in the stock in our base case given what this implies for FCF and reinvestment in the interim period.

Here’s a visual representation of what annual comps could be, ex-COVID, in ‘20E & ‘21E base case bridge to our 2021’s exit-rate productivity of 90%:

Risks

-          Faster productivity recovery

o   While we believe that our structural reasoning is logical and recent trends supportive of our view, we acknowledge that some of our assumptions are speculative and that we are extrapolating future trends from relatively few reopened stores over a short window of time.

-          Activist/strategic investors intervention

o   KKR owns a significant stake even after dilution which it had increased in March. However, their increased involvement came before it became apparent how much trouble PLAY is in. It is telling that KKR chose not to act when the stock was at ~$5. We believe that this risk is relatively small unless we are also significantly off base on the rest of our thesis.

-          COVID vaccine and headlines

o   An effective and widely administered vaccine could significantly reduce consumer hesitancy to return to arcades. We would point out, however, that a lot of optimism about COVID recovery is already priced into the stock alongside the market as a whole. COVID-related trends and headlines could prove to be negative catalysts as easily as positive ones. Furthermore, a COVID vaccine several months from now would not restore D&B’s missing capex, nor would it undo the investment consumers have made in at-home entertainment

----------------------------------------------------------------------------------------------------------------------------------------------------------------

Appendix

Comparison to leading casual dining chains suggests PLAY’s post-dilution valuation assumes that its future risks are not elevated above the rest of the group

Google Maps data from Friday night (June 5^th) further supports our customer traffic work

Physical layout and desired consumer experience make hygiene and social distancing particularly difficult

FY18 – FY19 New Product Launches

On their 1Q18 earnings call, an analyst asks, “any insight [into] your level of confidence that the virtual reality is going to be a material driver of traffic” to which CEO Brian Jenkins responds “we're very optimistic and excited about the VR concept that we're introducing, particularly with the intellectual property around Jurassic World. It's just a great launch of a platform that we plan to utilize with other titles over time. So it does represent the biggest investment we've made as a company in amusement single game. We're going to follow it up later in the year with another game that we think also resonate well with guests... And we view it as an attraction that can not only drive traffic that's going to look – it's going to present well on our TV spot, but also drive some incremental spend while they're in the store, because we are charging a separate price for it.”

Research methods

Throughout this report we make references to our various aspects of our alternative data ensemble, which we use to make operating performance projections and develop insights into fundamental trends. This ensemble includes over a dozen data sources which track the operating performance of several hundred consumer/retail/eCommerce companies. It includes multiple independent credit/debit card panels, geolocation panels, BillPay statements, web scraping, social media aggregation, eCommerce receipt panels, clickstream panels, google trends, and price/discount monitoring. In addition to leveraging our data ensemble, our research into PLAY includes consumer survey analysis and telephone channel checks.

Fwd EV/EBITDA during 2019

For reference, from Bloomberg, here is 1yr & 2yr forward EV/EBITDA throughout 2019 (6.7x & 6.3x on average across the year, respectively). We note that 2H’19 traded at a discount to 1H’19, given deteriorating performance, but chose not incorporate this penalty.

Disclaimer: At the time of publication, the author of this article holds a short position in DAVE & BUSTER’s ENTERTAINMENT (PLAY).  This article expresses the opinions of the author. The author has no business relationship with any company whose stock is mentioned in this article.

The author of this article has a short position in the company covered herein and stands to realize gains if the price of the stock falls. Following publication, the author may transact in the securities of the company, and may be long, short or neutral at any time.  The author of this report has obtained all information contained herein from sources believed to be accurate and reliable.  The author of this report makes no representation, express or implied, as to the accuracy, timeliness or completeness of any such information, or as to the results to be obtained from its use.  All expressions of opinion are subject to change without notice, and the author does not undertake to update or supplement this article or any of the information contained herein.  This is not an offer to sell or a solicitation of an offer to buy any security.

Catalyst

-        Upcoming Q1 and Q2 earnings reports should shed light on D&B’s recovery trajectory in both absolute terms and relative to restaurant and amusement peers.

-        Management elected not to take questions during their last earnings call and announced an equity raise shortly thereafter. If management again refuses to take questions on their upcoming Q1 call we would view it as supportive of our thesis.

-        Another liquidity raise near term is possible if management extrapolates current trends relative to Q4 liquidity needs and decides to take advantage of its recently elevated stock price.