Description

*Numbers in € unless stated otherwise as HSW reports in €. HSW is listed in the London and Irish Stock Exchanges quoted in £ and € respectively. 

I recommend an investment in Hostelword. With ~35,000 properties globally including 14,000 hostels listed in its platform and more than 7 million annual bookings HSW benefits from strong network dynamics and is the preferred choice for budget travellers as a result with ~ 40% share of bookings in that niche.

The market has apparently good reasons for not giving the company the credit it deserves: i) the stock is not optically cheap by any metric so it screens poorly, ii) the sell-side overestimates competitive forces by getting some industry-specific metrics wrong, iii) the market underestimates operating leverage out of marketing investments and iv) the accounting treatment for some items and especially for “amortization of intangibles” obscures HSW’s true earnings power.

Like Priceline and Expedia, HSW is an OTA. Unlike Priceline and Expedia, HSW only lists hostels and budget accommodations. A brief look into the past. Late 1990s and early 2000s saw OTAs disintermediating hotels and making tons of money out of it. As the growing travel market is large and benefits from attractive economics OTAs have attracted lots of competition. Metasearch companies have launched initiatives to bypass OTAs by offering the possibility to book directly in their sites – i.e. TripAdvisor has launched Instant Booking with little success so far. On top of that large hotel chains have refined their online channel to make the booking process more visible and friendly to customers. Incidentally the Hiltons, the Marriots & Co. still retain some bargaining power over OTAs as their penetration is high – in US large chains account for 70% of hotel bookings and 30% in Europe. So this decade OTAs face different degrees of risk of disintermediation themselves.

On the other hand “large” hostel chains comprise only ~ 13% and 8% of the market in the US and Europe respectively. You know Equity Point, right? Nor do I. The fact that Equity Point is neither a private equity firm nor a hedge fund but the hostel-equivalent of a Hilton tells you everything you need to know about the value of the brand when it comes to hostels. Comprising a long tail of independent properties sometimes under a sole owner-operator umbrella, the hostel market is vastly fragmented and its players would be individually irrelevant to end customers had not been listed in an OTA. So it is not surprising that the most active hostel chain on HSW’s platform only makes for <2% of sales. HSW bears little risk of disintemediation from hostels.

While HSW operates in the broader budget segment ~ 90% of its sales come from the cheaper hostel vertical. So essentially 90% comes out from 14,000 hostels and 10% out of the remaining 21,000 sites. This is not surprising as the high-end is more competitive with Expedia and Priceline leveraging their unrivalled amounts of inventory. However I think HSW is better positioned for the hostel segment. Expedia entered the hostel vertical in 2014 with not so much of a success insofar whereas Priceline just commands half of HSW’s bookings despite offering ~ 14,000 hostels in its platform, as many as HSW does. I think there are two main reasons for Priceline’s trail. First I find it just difficult that a company takes a segment seriously and invest accordingly when it makes for less than 1% of its sales. More importantly HSW has more inventory breadth and depth. Depth because even in big cities young travellers are looking less for a private room and more for a bed in a shared room. Breadth because it is not really the inventory in large urban areas where there are many alternatives what matters most for long duration travellers but the inventory available in minor cities and remote locations. As most HSW’s travellers are young, go solo and ~50% travel for more than two weeks I think HSW’s dorm offering in unconventional locations gives the company an edge over Priceline in that niche.

Incidentally HSW bought in 2003 and 2013 its two pure-hostel competitors Hostels.com and Hostelbookers.com, which remain as independent brands but have a minor and shrinking contribution to HSW’s revenues. These two acquisitions left HSW as the only pure-hostel OTA player with a truly large scale. Concerning the AirBnBs, the Homeaways and similar newcomers of the sharing economy none of them to my understanding can parallel the social aspect of hostels. So unless you disagree with Aristotle on the fact that men are by nature social animals and/or you lacked hormones in your twenties, you know what I am talking about. https://www.youtube.com/watch?v=w0xbODE6VLc

So with limited competition, what would it take for HSW’s users to migrate to a new platform? More inventory. Better prices. Or both. On inventory, it is unrealistic to think that a new entrant will capture significantly more than HSW’s current ~14,000 hostel properties with some estimates of ~18,000 hostels worldwide. Actually chances are that HSW captures most of those remaining ~4,000 over time. Just a clarification on prices. With some exceptions, there are mainly two types of network platforms. On “two-sides pay” platforms like AirnBnB, users pay a fee for rentals and the company also gets a fee from service providers. On “one-side pays” platforms one side subsidises the other. On Google and Facebook advertisers pay but the service is free for users. Controlling for inventory I think it is obvious the one-sided model is more sustainable. Nobody can compete on price grounds with a free model for users. HSW follows a one-sided model.

HSW charges a fixed commission to hostels on every booking ranging from 12 to 15% depending on the type of accommodation and location. Not different to the ~15% Priceline takes on average. On top of that hostels listed on HSW also have the “Elevator” option, bidding a premium commission rate up to 800 bps in return of a higher ranking on the search results page within Hostelworld. So in some cases up to 23% of the booking value goes to Hostelworld. That is a lot of real cash especially given that HSW charges its clients upfront through a non-cancellable deposit equivalent to the commission rate HSW´s charges to the hostel. Therefore from a monetization perspective for HSW it does not matter whether after booking the client shows up in the hostel or not.

Why would a client accept to pay a non-refundable amount when other platforms as Priceline allow for charge-free cancellations? Why would hostels take less than the whole value upfront when most OTAs allow them to take 100% and pay the commission at the end of the month? On the client side I think it there is some sort of framing bias in place. As half of HSW customers book within the same week of arrival it is just difficult for them to mentally discount that an unexpected event might ruin their vacation. Indeed a bunch of clients might book during the actual travel, so despite other booking options might be just one click away it is just inconvenient to look for alternatives other than the one that provides them with more affordable choices. Also consider again that Priceline and other platforms do not include many secondary locations long-time duration travellers seek.

On the hostel side I find two main reasons for choosing HSW as a platform. First, other than some little pain for hostels managing bookings there is no direct economic cost for operating with several platforms at the same time as generally OTAs charge for booking and not for listing. So I think it is safe to assume the more listings the more revenue. There is also a conversion factor to bookings. If a traveller has already committed part of the booking price in the form of a non-refundable deposit he has simply more incentive to show up, so bookings on HSW are more likely to convert into actual bed nights. Indeed while Priceline gets around 40% of its bookings cancelled HSW has just 7% and less than 10% of no show-ups. Getting cash upfront and having minimal maintenance costs for the platform - no-growth capex averaging <3% of sales -  make the company very cash generative with ~90% of EBITDA turning into free cash flow.

Unsurprisingly given its great underlying economics net cash on the balance sheet represents ~6% of HSW’s market cap so essentially the 2015 IPO was not really a capital raising effort to expand the business but a way for the previous main owner, the PE firm Hellman & Friedman, to cash out. Despite an IPO might not give the highest exit value the Irish tax law prohibits to give direct shares to a new owner so going public was the most sensible option. This tax-legal incentive to be a public company in combination with the company paying dividends reduces the chances of a classic nightmare: asset-light and cash generative businesses raising money they do not need to grow but to fulfil management desires of lavish offices, jets and other perks. I think this is not the case.

More worrisome for the market is the fact that after years of positive comps, 2016 sales and profits compare poorly to 2015. Let put those numbers into perspective though. Sales are driven by: i) bookings and ii) average booking value (ABV), which is basically the commission rate HSW charges to hostels times the final booking price for the client.

2016 bookings were essentially flat in vs. 2015. I think there are two main reasons for the slowdown. 2016 has seen all time high awareness on the threats that terrorism poses on our lives. While 2015 saw the infamous Paris’s Bataclan night nothing compares to the social stigma that the 2016 Nice and Brussels episodes have fuelled. While the actual impact on bookings is impossible to measure Europe has logically been the most penalized region with bookings falling by 10%. To make it clear if for any reason people stop traveling in mass for long this thesis will not work, yet I personally think that either feeling relatively safe or not that much people end up travelling.

The second reason is company-specific, a shift in brand and marketing investments. Indeed, it seems like a plausible explanation for the slowdown that after spending €37m or 45% of its sales in marketing in 2015 and just €33m or 41% in 2016, HSW is currently underinvesting somehow. I think these numbers are misleading though. HSW still deployed in 2015 some resources not only for its flagship brand Hostelword but also for Hostelbookers.com and Hostels.com. Now the company has repositioned to direct its branding and AdWords bidding efforts to Hostelworld. This seems validated as only 13% of bookings now come from secondary brands vs. 27% in 2015. However shifting bookings to the main brand is an asymmetric game in the short-term. The loss of bookings out of underinvesting in the secondary brands is not fully offset by the gain of bookings in the main brand. The reason is simple. No matter the brand awareness you create at the beginning your company will get less visits if it still lies in the first place on AdWords results but has disappeared from the second or third places in most cases. Transition to mobile has emphasized this phenomenon as smartphone users care less about the 4^th or 5^th option in the list than those on desktop. Needless to say this brand bypass is temporary and as soon as it ends comps will be more meaningful.

In addition to the impact on growth of terrorism and marketing shifts there is a structural reason that limits growth. Because of the fragmented nature of the industry, online penetration for hostel bookings is ~70% vs. 40% for hotels. It is comparatively more difficult to make a bull case of bookings transitioning from offline to online.

However there is one factor I think outweighs the “non-growth” arguments. The opportunity to dramatically expand the client base by changing their current perception on hostels, especially in the case of millennials as the sort of public that hostels might appeal to. Phocuswright, a relevant travel research agency, estimates that just 15% of millennials stay often in hostels and recognize their real benefits. What about the remaining 85%? Well, while there are 3 out of ten that would never consider staying in a hostel, there are ~55% that would consider a hostel as a serious option but have NOT really used the service so far. This is understandable amid historical perception of hostels as crappy and unclean places where also some of your dollars might mysteriously disappear from your room without previous notice... I think this perception is changing notably as hostel chains and particularly HSW are emphasizing aspects like the option of private rooms and spending notably in TV ad, which has proven to be very effective insofar to build awareness. Indeed, that 70% online penetration for hostels bookings bodes well for more awareness translating into bookings, bookings in the form of online bookings, and online bookings in the form of HSW’s bookings. Look at 50 cents: https://www.youtube.com/watch?v=M-dfdZJafFo

Concerning ABV some sell-side reports highlight the 2016 drop to €11.6 in vs. €12.1 in 2015 providing what seem to be sensible reasons for the decline and extrapolating a long-term trend out of it. They will tell you that the Elevate Penetration rate and the average effective commission rate are at their peak historical levels at 33% and 14% respectively. They might also tell you that booking a dorm in Vietnam and more broadly in APAC, the fastest growing region, costs significantly less than one in Europe or the US. They might also show you that transition to mobile, with more than 50% of current bookings and growing, is accelerating pressure on ABV as most bookers through that channel travel solo as opposed to desktop where group-bookers are more relevant in the mix.

While I think all those arguments are true they are incomplete. For one 2016 was a terrible year for bookings in Europe, an aberration, devaluing long-term trends that use 2016 as the base year. In other words the cheaper Asian accommodations won’t gain nearly as much weight in the mix as it seems. Second the rapid shift to mobile has  positive aspects that those arguments miss. Smartphones’ screen size incentivizes bids for better positioning on Elevator as few people care about the 20^th hostel in the list, so I believe Elevate penetration and its related commissions are unlikely to experience any sort of mean reverting dynamics. The fact that ABV for solo-bookers is lower is true, but it is not less true that solo bookers book more times on a trip. Less ABV offset partly by more bookings. So I do not think ABV trend poses as much pressure on margins as some analysts think.

Phocuswright projects 7% - 8% hostel revenue growth per year through 2018. I have modelled growth more conservatively at 6%, despite HSW as the consolidator should grow even more aggressively than the industry as a whole. Concerning brand and marketing I think 40% of sales is actually a conservative number for two reasons. First what I already stated about the discontinued investments in supporting brands makes it unlikely to reach 45-50% plateaus. Second is the effect of HSW’s brand awareness upon TV and SEO investments for years. Every day that passes more people book directly, click-on free organics results or download HSW’s app, with currently ~60% of bookings coming from unpaid traffic. More importantly and to calm down those who still believe that regardless only 40% of bookings come from paid-traffic AdWords inflation will continue to be a drag, I would argue that paid traffic is still EBITDA positive and has positive side-effects on organic results. Indeed cost-per-click inflation on Google has decreased over time for HSW, with 2014 and 2015 YoY increase of 40% and below 18% respectively. I do not have actual numbers for 2016, but my best guess is that number is now much lower and decreasing. The reason is that as more and more people become familiar with the brand it is simply less costly for HSW that users click on paid links related to searches of  “Hostelworld” rather than “Hostels in London” or “Cheap accommodation in San Francisco”.

As WEB states in his 2015 and 2017 letters, there is the type of amortization that is a true economic cost like software that needs to be up to date so it can compete and there is the sort of amortization that is only a pure accounting item with no economic substance behind. I think the latter is the case with “domain names” in HSW that are distorting the income statement. Hostels and Hostelbookers were acquired with the main purpose to accelerate consolidation and to redirect traffic but not with the intention to keep them as going concerns. So in the 2016 P&L there are ~€8m impairments and ~€13m in amortizations that can be added back to income safely. Adjusting for €1m recurring capex and €2m from capitalized costs that I expense, I get ~ €19-20m free cash flow, which is 5-6% FCF yield ex-cash on a no-growth state. Adding the conservative top-line growth assumptions and some operating leverage at the G&A and marketing levels I get to €24-25m of free cash flow in 2020, which is 6-7% CAGR. So IRR would be 5% FCF + 6-7% FCF growth = 11-12% assuming no multiple expansion.

PS: Sorry if 12% is not great for you but I think it is a good return given the investment opportunity set and given that I have discovered a new type of duration mismatch in addition to the classical asset-liability mismatch. The timing mismatch of generating an idea, researching it and posting it on VIC sometimes translates into a significant difference in price. When I started this write-up the company was 10-15% cheaper...mmm…. Anyways I think it still is a good opportunity.

Catalyst

- Client base growth 

- Share gains from smaller comps and hostel chains

- Acquisition by a larger OTA?