Following the fourth-quarter 2015 earnings release, 2015 10-K and a call I had with CFO Peter Hovenier, I wanted to write a follow-up article on my favorite holding, Boingo Wireless (NASDAQ:WIFI). Rarely do I find a stock interesting from both a modelling and story perspective.
Boingo has made an impressive pivot from providing retail Wi-Fi hotspots in airports and other venues to becoming one of the largest installers of distributed antenna systems and Wi-Fi networks. Going forward, it will offload cellular carrier data onto its wireless networks completely transforming Boingo as a company. At a current price of ~$7, I see 100% upside based on a 2016 year-end discounted cash flow valuation of $14.66. Given the direction this sub-industry is moving, I feel like the chance of "tail-risk" is more limited than in other companies, i.e. the chance of the business having limited value in perpetuity is low especially given the infrastructure it has built out.
I already did a briefing on Boingo segments in my previous article, but I wanted to re-summarize its segments with a refresh on strategy and numbers.
Boingo Wireless provides a global network of Wi-Fi and distributed antenna systems (small cells instead of big towers to "boost" cell service in highly concentrated venues. Boingo partners with airports, arenas, colleges and other highly populated, data-intensive areas and provides both Wi-Fi infrastructure and DAS (partners with cell carriers for this). As one of the few independent companies that can provide both DAS and Wi-Fi, Boingo has been very successful in winning new venues.
The company organizes its revenue into the following categories: retail, military, DAS, wholesale, advertising and other.
The company allows consumers to purchase single-use or subscription plans to access the Internet at Boingo's managed and operated hotspots and partner locations. This was the company's heritage source of revenue and had very high incremental margins. Airports and other locations have moved away from this subscription model and instead are providing free or freemium Internet. Freemium Internet is Internet provided for a short period of time (e.g. 30-90 minutes) with the option to pay after the time limit. As venues have switched to the free/freemium model, retail revenues have fallen significantly and likely will continue to fall going forward. I expect revenues in this division to fall off a cliff as subscribers drop very rapidly and single usage essentially goes away. Losing $30+ million of a high-margin business would be a difficult pill to swallow except for the new segment it has just about finished building out...
Recently, Boingo has entered into the military broadband IPTV market with a significant capex over the last few years. It has built out 244,000 beds (as of 12/31/2015) out of a potential 300,000 beds with 2015 revenues at ~$20 million. By working across multiple military locations, users can seamlessly transfer their account as they move locations. Additionally, at a military base, there is no installation and no equipment required. Personnel need only their laptop or device and can use broadband Internet and access IPTV. Boingo will finish the remaining 56,000 beds with $15 million of military capex in 2016. I estimate revenues will be roughly $30 million in 2017 and stable from this point. It does not have any plans for additional beds, but once the company's cash flow stabilizes, that could be an area of growth. Early results have penetration and churn performing better than expected.
Distributed antenna system, or DAS, is one of the fastest growing and most important segments for the company. Boingo installs small antennas in venues (vs. the traditional large towers) to improve service in high density, high usage areas. Major cell carriers have already started making the switch to small cells, and this trend could continue going forward. The venues will typically select Boingo to install and manage both the Wi-Fi and the DAS systems in a venue. Unlike a tower company, Boingo can handle both of these responsibilities and unlike one of the cellular carriers, the company is independent. It is extremely frustrating to be at a concert or sporting event and be unable to send a text or snapchat a video. Improving service in these areas is a key point of focus for the carriers. Typically, Boingo starts with partnerships with 1-2 cell carriers. These carriers lead installation and pay for all up-front capex. Additional carriers typically sign on at a later date. On an ongoing basis, Boingo receives access fees to manage and service the system. It makes a margin during installation, recurring access fee revenue, and upgrade revenue (i.e. 3G to 4G).
DAS revenue grew 21% in 2015 from ~$38 million to $46 million. Management has indicated build out versus access fee revenue is split 60%/40%, increasing to access fees over time. Currently, it has 10,900 live DAS nodes with 4,900 nodes in the pipeline (agreements signed but nodes are not yet completed) so the growth prospects are clearly strong.
Wholesale is where I believe the real future of the company lies. Wholesale traditionally has been associated with partnerships with other companies (like cable companies and American Express (NYSE:AXP)) that allow their clients and users to use Boingo for free as a benefit. Now wholesale includes cellular offloading agreements.
Sprint (NYSE:S) has signed the first offloading agreement with the company whereby its users will automatically divert data use to Boingo hotspots when in range. This transition will be completely seamless from the users' perspective and will reduce the strain on the carrier's network. In return, Sprint will pay Boingo a fee per MB used. It does a great job of explaining offloading and the economics in this transcript. It has successfully merged most of Sprint's subs by the end of 2015, and based on management commentary during recent earnings calls, it should be announcing a second major carrier anytime now. I assume it will be T-Mobile (NASDAQ:TMUS), but if it is AT&T (NYSE:T) or Verizon (NYSE:VZ), expect to see the other big boy join on quickly.
The exciting thing about offloading is this could be a $300 million business for Boingo with ~50% EBITDA margins. For a company that does ~$160 million in revenue and ~$40 million in EBITDA ('16 guidance), this kind of growth would be game changing. Management continues to state that it is a when, not if question of getting all four carriers on board. There are still larger questions in terms of pricing and how much data actually gets offloaded. Boingo expects carriers to be selective about when they offload and when they use their network. During non-peak times for example, it would make sense not to offload as their network can handle the bandwidth. During peak times, however, the cell company would likely offload as much as needed to keep overall network speed to all users relatively high. It will be a delicate game the carriers have to play as they manage user experience with expenses paid to Boingo.
Boingo provides advertisements when consumers log into their wireless networks. Some venues have consumers watch a short ad to access the network. These revenues had very strong growth the last two years, but have shown weakness in 2015. I would assume this segment could have more significant growth going forward, but given the 2015 performance, have kept growth muted at low-single digits.
Please download my latest version of the model here. The first tab is a basic comp based on the latest 10-K. At current price of ~$7, Boingo has a market cap of $274 million and a firm value of $283 million.
The second tab is a revenue build by segment, cost build, and DCF.
Retail shows significant weakness with a total decline over 20% in 2015. I assume this continues going forward with single-use continuing to fall by 35% per year and subscription tapering off as gross adds fall off and churn degrades the subscriber base. Boingo no longer breaks its retail revenue into pieces, so I make assumptions on the single-use and subscription revenues.
The new military segment makes up for much of the loss of retail. In 2016, installed beds rise to 300,000 and penetration stays in the low 20%s. I assume churn is flat at 3% and ARPU growing at 2% per annum. This gets me to a roughly $30 million run-rate revenue.
DAS has shown strong growth the last two years as Boingo has had many venue wins. Near the bottom of the model, I have built out the DAS revenue using a combination of historical information and inferences based on management commentary of DAS split between access fees on existing venues and new builds. I assume node growth slows significantly (management has said the growth over the previous years is exceptionally high), but that access fee revenue grows at a rate of 4% per year as I smooth out the upgrades that will inevitably take place (LTE+U, 5G to name a few on the horizon).
Wholesale, ex-offloading, is expected to decline as several partner contracts roll-off. The benefits of giving customers free Wi-Fi are not as valuable as it once was, thus these partnerships are waning. However, there are new partnerships to be had as Boingo has teamed up with Time Warner Cable (TWC) (potentially the new conglomerate) in a Wi-Fi sharing arrangement.
Advertising had a poor year in 2015 despite the increase in free Wi-Fi. I think the longer-term trends for this division should be positive. I have only assumed a 2% growth rate going forward given the headwinds in 2015. I think this is a point of conservatism; we have seen how big advertising can be for many tech companies once they figured out monetization. With 100% free Wi-Fi an eventuality for Boingo users, it will have a lot of eyeballs over its networks. It should be able to figure out some type of banner ad or targeted one-time ads to juice the advertising revenues.
As mentioned previously, a big part of the business model for Boingo going forward is wholesale cellular offloading. Towards the bottom of the model, I built out this portion of the business separately from everything else. I have inserted comments with each of the line items to explain their significance. Boingo's management has repeatedly stated that it is a matter of when and not if it signs all four major carriers. I have assumed this happens slowly over the next 15 years. Faster adoption and penetration of the carriers' customers will only increase the value of Boingo. I assume penetration stays at 55% as cellular companies won't offload all customers' data to Boingo at all times. During non-peak times for example, it is likely that Verizon will use its extensive network to handle data; only offloading customers when required to maintain its normal high speeds.
I assume customer usage per session starts at 30 MB and grows at only 5% per year. I think this is another area of conservatism. I assume the price per MB charged is ½ cent and remains constant. I have been struggling with this assumption as I wonder if the pricing will come down. However, by keeping the price constant in nominal terms over 15 years, that is a significant decrease in real terms. Management guided to an EBITDA margin on offloading in the "high 50s, low 60s" but I assume 50% margin.
If you scroll to the cost side (above), you can see I assume decreases in costs as a percent of sales across the board. A big reason for this is that many costs associated with the growth of Boingo (offloading) are already incorporated into this 50% margin. I assume the stock-based compensation more than doubles, however, as SBC is typically excluded from management's calculation of EBITDA.
The final result is a year-end share price of ~$14.66.
There are a few major risks for Boingo. I believe they can be mitigated, and as you'll see in my downside model, they have been mostly incorporated into the current stock price.
- Inability to win new DAS/Wi-Fi venue contracts.
- If Boingo cannot continue to increase the size and breadth of its network, the company will not only lose out on DAS revenues, but also may not be seen as "must have" for cellular clients regarding offloading.
- All four cellular carriers do not sign up with Boingo/terms of offloading changes significantly.
- If one of the large players (Verizon or AT&T specifically) does not sign an offloading agreement, Boingo's revenues and profits would be significantly affected.
- Would also provide leverage to the other carriers to negotiate more favorable terms.
- Reduced revenue per MB or lower data offloaded would also have a negative effect on Boingo's financial results.
- During Boingo's pivot over the last few years, expenses have increased significantly. If it is unable to control expenses, Boingo's financial results and share price could suffer.
In order to quantify these risks, I created a secondary model with downside assumptions.
- Assume lower DAS backlog and lower access fees.
- Assume network access, operations and sales & marketing remain constant as a percent of non-offloading sales from 2015.
- Assume development & technology nominally grows by 1% per year going forward.
- Assume Boingo loses out on AT&T or Verizon and only has 74% carrier penetration.
- Given these assumptions, I estimate a 2016 year-end per share value of $6.25.
For me Boingo is the perfect storm - a simple one-business company that is relatively easy to model coupled with a strong story that is easy to understand. It may take several years for the market to "properly value" (in my opinion) Boingo. In 2017, military capex will be near zero and will likely have two carriers offloading. This should accelerate free cash flow generation and allow Boingo to initiate a dividend or large share repurchase that will likely lead to improved perception in the public markets.
Disclosure: I am/we are long WIFI.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.
Additional disclosure: I may buy or sell shares at any time.