The Long Case For Liberty SiriusXM

Summary

• We view SIRI as an attractive monopoly in the satellite radio industry with strong organic unit growth, significant operating leverage, and low capital intensity.
• Mispricing exists because the Street overplayed fears of competition from music streaming and unjustly written off business’ long growth runway, especially in its used vehicle segment, as auto cycle peaks.
• The market has also overlooked significant option value in the monetization of valuable spectrum assets and a continuation of aggressive share buybacks by an incentivized management.
• There is an opportunity to invest via LSXM.K, a pure tracking stock for SIRI that trades at a sharp 17.9% discount to NAV, which we believe is unjustifiable and will collapse due to a number of visible catalysts.
• Our conservative base case scenario significantly discounts the realizable spectrum value, largely excludes optionality of levered share buyback, and values LSXM.K at $57.48, which presents an asymmetric risk/reward upside of 44.9%.

Investment Thesis

We view SIRI as an attractive monopoly in the satellite radio industry with strong organic unit growth, significant operating leverage, and low capital intensity. Despite a formidable core business, we believe mispricing exists because the Street has overplayed fears of competition from music streaming and unjustly written off the business’s long growth runway, especially in its used vehicle segment, as the auto cycle peaks. The market has also overlooked significant option value in the monetization of valuable spectrum assets and a continuation of aggressive share buybacks by an incentivized management, which provide additional pathways to unlock value. Augmenting this attractive core investment, there is an opportunity to invest via LSXM.K (NASDAQ:LSXMK) , a pure tracking stock for SIRI that trades at a sharp 17.9% discount to NAV, which we believe is unjustifiable and will collapse due to a number of visible catalysts. Our conservative base case scenario significantly discounts the realizable spectrum value, largely excludes the optionality of levered share buyback, and values LSXM.K at $57.5, which presents an asymmetric risk/reward upside of 44.9%.

Value Driver I: Overblown concerns about music streaming have distracted the Street from SIRI’s unshaken value proposition and entrenched monopoly.

The Street’s concerns about the rise of streaming services misunderstand SIRI’s value proposition to its core consumers. SIRI operates as the only satellite radio broadcaster in the U.S. offering a wide selection of ad-free, high quality music and talk radio stations, with a 75% penetration rate on all new cars sold in the US. The Street incorrectly views streaming services such as Pandora, Spotify, and Apple Music as direct competition to SIRI for multiple reasons. First, this concern fails to take into account the nature of SIRI’s older, affluent subscribers, who are car owners and have an average household income at $110,600 vs. a nationwide average of $72,641. With channels devoted to Bruce Springsteen and Elvis Presley, SIRI’s typical consumer is middle-aged with a long commute, who would quickly get bored listening to the same Spotify playlist in the car every single day. SIRI’s value proposition focuses on delivering high-quality and ever-changing playlists painstakingly created by talented hosts. This ‘lean back and relax’ radio experience is differentiated from the self- curated listening experience of streaming apps, which the older population has little time for. Millennials may be more active listeners, but when they age and face commutes and working lives - incidentally when car ownership rises - passive listening makes more sense. Incidentally, in the US the average age of new car buyers is continually rising and was at 48 years old in 2015. Additionally, SIRI offers more than just music. SIRI offers high-quality coverage of MLB, NBA, and NFL programming which cannot be offered by these streaming competitors. Management has confirmed that their $15.99 Sirius Select plan has the most subscribers, and SIRI’s paying subscriber ARPU of $13.6 (understated due to legacy plans/discounts) represents this mix, implying that most subscribers are willing to pay an extra $5 for non-music content. Second, SIRI subscribers are customers who know their options, and have chosen SIRI over/with competitors’ offerings. 40% of SIRI’s subscriber base has used Spotify at some point, and more than 60% has used Pandora. Subscriber counts for SIRI (22.6% increase since 2013) have also risen in tandem with these services (Spotify: 500%, Pandora: 33%) pointing to significant market segmentation and growth in overall music listening hours. Tellingly, even as streaming has grown SIRI’s churn rate has not budged an inch, as we would expect if there was direct competition. Third, streaming has low switching costs as well as margin pressure from large-scale entrants with external monetary support, leading to an unsustainable strategy of discounts and promotions which will make current pricing and subscriber growth short-lived. Spotify’s effective monthly retail pricing, for example, has come down 26% over the last two years, compared to SIRI’s 5% ARPU increase over that same period - we believe this pricing power is further evidence of market segmentation. For radio listeners, current mobile data prices make it infeasible to listen to radio streaming through Spotify or Pandora (without pre-curation) as a substitute, as customers would have to pay $10.0/GB, which for an average commuter would cost $12.8 in data costs per month on top of a Spotify subscription fee.

SIRI is resilient to industry-wide pricing pressures due to monopolistic business and advantageous cost profile. SIRI is particularly well-positioned compared with traditional terrestrial radio for many SIRI for many reasons. First, terrestrial broadcasters tend to have over-levered balance sheets which will constrain their access to capital which they need for further growth -- seen in terrestrial radio’s flat annual revenues since 2011. Second, streaming competitors have seen margins compress due to rising music rights costs(which are variable to subscriber count), while SIRI has much of its programming costs locked in for several years. Content costs as a percentage of sales, for example, are 17.5% for SIRI, 53% or Pandora, and 84% for Spotify. More importantly, this is indicative of the enormous scale that SIRI enjoys, where they are able to spread a comprehensive content package over a huge paid subscriber base of 24.3mm v. ~10mm US paid for Spotify. Equally important is that we expect only a minority of the streaming subscribers would pay extra for non-music content, leading to lower effective scale for that content. This advantageous cost structure lends to SIRI’s immense operating leverage with 70+% incremental contribution margin and the content distribution industry’s highest EBITDA margins at 35%. Third, SIRI’s business is exceptionally capital-light with 2018E run-rate unlevered free cash flow conversion at 93.3% (2016: 96.9%, highest in industry). The high conversion rates partly due to the Company’s $1.2bn in NOLs which will nullify any cash taxes until roughly 2019. Capex as a percentage of sales has also been stable at <5%, while ROIC has grown from 8.3% in 2011 to 13.7% in 2016.

Extensive dealership network, aligned OEM incentives: SIRI’s dealership network of ~37,000 new and used car dealers has been its main distribution channel. On top of this high barrier to potential competitors, SIRI has strategically aligned its incentives with players across the entire value chain, reaching OEMs. Car manufacturers have been eager partners and are kept happy not only through simple radio subsidies, but substantive revenue share agreements on subscription revenues arising from installed radios until the end of the car’s life. We view this as a huge competitive advantage. On the surface level, OEMs are incentivized to install the radios in such a way to facilitate SIRI’s subscriber conversion, in order to maximize their revenue share. Less intuitively, OEMs have actually been proactively trying to improve SIRI’s competitive position. In the near term, OEMs have incorporated SIRI in their planning process on next-generation connected vehicles stretching to the 2022 models. Effectively, SIRI is part of their long-term vision and has the ability to adapt to future cars designs as they are in the loop. Dealers have been helping SIRI through the Service Lane Program, which reminds drivers to renew their subscriptions or activate a trial/new subscription every time they come in to the shop for repairs. Both dealers and OEMs have been sending SIRI very comprehensive data on subscribers and sales transactions, including name and address, VIN, make of car, even ethnicities.

Value Driver II: SIRI’s growth runway in an underpenetrated used vehicle market and above-potential churn rates are obscured by new vehicle market fears.

SIRI’s Used vehicle segment is highly underpenetrated and represents the next frontier for growth. Consensus views on SIRI’s organic growth do not take into account the approaching penetration wave of SIRI radios in used vehicles. This is due to a lack of understanding of the dynamics of how used vehicles with radios enter the used vehicle market. Gross add figures given by sell-side analysts incorporate strong pessimism on the state of new car sales due to a much-touted surge in off-lease used vehicle supply in 2017/18, but fail to take into account the largely untapped market potential present in the used vehicle market. We incorporate this in our top-down model, building in Manheim projected off-lease supply growth and slight (~0.3%) growth for other used cars sales due an increase in their affordability. We project the penetration of SIRI radios in used vehicle sales rising from 31% in 2016 to 58.8% in 2021 for several reasons. First, we note that on average new vehicle owners tend to own their vehicles for 7-8 years before selling them to another consumer, with this average lifespan indicative of the new layers of cars that will be entering the used car market in 2017-2018. The impending entry of these cars into the used market is a result of the high new vehicle penetration rates that SIRI first achieved in 2009-2010 with the signing of contracts with OEMs including Toyota and Ford, making SIRI radios standard on all their vehicles sold in the US. These layers are expected to flow into the used market in 2017-18, increasing SIRI penetration of the overall used pool of cars. Therefore, we anticipate a significant increase in radio re-activations in used cars, causing a shift towards used car activations as a percent of total customer additions. This shift will also serve to lower SIRI’s overall subscriber acquisition cost (SAC), as used vehicle adds do not require the cost of radio purchases and heavy OEM subsidies, making them much cheaper. While used car customers have lower conversion rates compared to new car customers, management notes that spending patterns after conversion are identical between groups, ensuring that this mix shift will not lead to decreased ARPU. We have begun to see this shift occurring through our channel check with used car dealers, who appear to have observed an accelerating penetration. Second, the impending growth of the used car market as a portion of total SIRI revenues will also see growth due to management’s promotion of satellite radio trials among dealers of used cars. SIRI began focusing on this avenue for increased reach in 2012, reporting the number of used car dealers with which they had made connections with - for the period 2012-2016, the number of used car dealers working with SIRI has grown at a 26% CAGR, reaching 26,000 dealers in 2016. The core result of this extended outreach will be a mix shift, with the implementation of new channels beyond traditional dealerships, such as service bays and insurance companies, re-selling vehicles will serve as a path towards increased used vehicle penetration. Further, by examining the extent of outreach that management has made historically, there is significant runway for growth - the reach of 26,000 used vehicle dealers implies that SIRI has an established reach in 37% of U.S. dealers, and conversion is bound to increase as SIRI’s outreach program matures.

Current churn rate is overstated due to a skewed contribution from involuntary churn, which should gradually decline over the next few years. We believe that there is sizable upside in terms of churn rate performance, which management guides at 1.8-2.0% LT. Total monthly churn is a combination of voluntary churn due to dissatisfaction (0.45%) and involuntary churn due to vehicle replacement (1.45%). The latter has been ticking up as the company grew, as a maturing portfolio with cars approaching the end of usage cycle, in comparison to a brand new fleet, is much more likely to see vehicle disposal and upgrade. As a result, the customer will not keep the previous subscription and be counted in churn. However, the company has been able to reduce enough voluntary churn to offset this headwind and consistently achieve a stable overall churn - illustrating the increasing stickiness of the product offering due to superior content addition. We believe that SIRI’s current vehicle replacement churn at 1.45% has almost converged with the industry average vehicle turnover rate. Looking ahead, this headwind will materially ease, as the company reaches a mature stage when average portfolio vehicle age becomes relatively stable. In addition, the average length of vehicle ownership in the US has been steadily increasing, prolonging the turnover cycle - ownership length increased from 42.0 mo. in 2005 to 66.2 mo. in 2015 (IHS Automotive). That translates into an 87 bps decrease of monthly involuntary churn (2.38% in 2006 vs. 1.51% in 2016). This 1.51% essentially caps the maximum normalized involuntary churn of a mature fleet. A lower involuntary churn translates to fewer chances for a user to leave the system, fewer 3-12 month promotional periods and lower reacquisition costs.

The economics of OEM promotional campaigns are not fully analyzed and understood by the Street. The company regularly discloses a 3-month new vehicle conversion rate from OEM-paid promotional subscribers, which is heavily watched by the sell side. This statistic has been trending down from 45% in 2012 to 39% in 2016. As a result, the sell-side is worried that the campaigns are becoming less effective in driving self-pay subscriber gain. At face value, this interpretation seems sound. However, this is a little bit misleading because starting in 2014 a number of OEMs - Ford and Chrysler - started to pay for customer’s longer trials at a fee to encourage conversion. This coincided with a 3% decline in the 3-month conversion rate in that same year. The explanation here is that customers with a 6 or 12-month trial period are less likely to immediately subscribe after 3 months and be counted into the 3-month new vehicle conversion rate. Consequently, this conversion statistic is artificially pushed down by the trial duration increase, which is actually positive for SIRI, since more time to sample the various programming and discover programming would better encourage users to become a self-pay subscriber.

Value Driver III: Substantial option value in spectrum value and share repurchases are unjustly written off by the Street.

Monetization of valuable spectrum assets has not been factored in. Following the merger of Sirius and XM, SIRI now owns two neighboring 12.5 MHz bands of spectrum through which they transmit their services to their customers. They have begun migrating their customers to one 12.5MHz band, leaving the other band open for other uses or a sale once the conversion to one band is complete. We believe the potential sale of this spectrum presents the possibility of additional upside due to the monetization of this band, particularly due to the FCC’s granting of 5 MHz guard bands which assures the wavelength quality. The Street has refrained from underwriting this spectrum value due to its longer horizon, but we think discounting conservatively addresses this problem. The saleable spectrum was valued using precedent transactions of similar spectrum on a MHzPop basis in 2025 for conservatism and discounted to present value.

Aggressive share repurchases amplify returns and catalyze value realization. Over the past four years, SIRI management has aggressively engaged in $7.95bn of levered share buybacks, of which $3.7bn was funded by additional debt, reducing share count by 8% for each of the past 5 years. Management has stated its commitment to continue levered share buybacks with a target net leverage ratio of 4x (2.96x currently), and a current authorization of $2bn which we believe will be renewed if not enlarged next year. As discussed above, we believe the market does not currently price in SIRI’s full potential and these repurchases will be significantly accretive. For conservatism, we have modeled in no additional leverage in our base and bear scenarios but have done so in our separate base levered buyback and bull scenarios.

LSXM.K Mispricing/Catalysts

The Street cites a number of reasons to try to explain the 17.9% discount of LSXM.K vs. SIRI. First, tracking stocks warrant some discount relative to an asset-based common stock. Second, unlike LSXM.K, SIRI is aggressively buying back its shares, which creates upward technical pressure on SIRI not present for LSXM.K. Additionally, whereas SIRI pays a dividend, LSXM.K receives the cash and has not distributed it - dividend-paying stocks currently command a premium in the market. Third, the market fears that Liberty might use LSXM.K as acquisition currency to buy SIRI shares that it doesn’t own at a premium. Fourth, the market is concerned about Liberty potentially acquiring Pandora which would complicate the LSXM.K holding structure, and for which Liberty could overpay. In a recent company conference presentation, sell-side research analysts polled an audience of large institutional investors about the Pandora question in which the voting results came out as, “Okay. So 25% said Siri buys it. 38% said Liberty Sirius buys it. 13% said another firm. And 1/4 of you think Pandora remains a stand-alone firm”. Clearly, the Pandora acquisition sentiment still exists in the market. Our belief in regard to each of the points above is: First, the LSXM.K holding structure is the least difficult of the Liberty tracking stocks for the market to value given its simple interest in SIRI and minimal credit risk at the parent level with only a $250mm margin loan. Second, the reduction in SIRI’s outstanding shares still benefits LSXM.K via a higher equity percentage ownership. Third, LBTY has explicitly ruled out the possibility of a discounted LSXM.K offer for SIRI shares and an acquisition of Pandora at current prices. Fourth, in their most recent investor call Liberty Media devotes an entire slide to the lack of profitability of streaming players, showing Pandora as having higher content costs relative to SIRI and placing an image of a gravestone next to failed streaming companies. We doubt that the legendary John Malone would viciously attack Pandora one day and then overpay for it the next.

There are a number of visible pathways to close the valuation gap. First, LSXM.K’s approach towards consolidation will be noticed by the market as the date nears. The Street currently estimates that LSXM.K will reach 80% consolidation in 2020, and as a result heavily discounts or does not focus on the value realization benefits offered by this consolidation. In our levered buyback case, LSXM.K is set to reach this consolidation point in 2019, and as this becomes apparent to the Street, the value gap will rapidly close by YE2018. Consolidation will serve as a major factor in the stock as the 80% threshold allows for Liberty to divert SIRI’s free cash flow upstream to LSXM.K in order to buy back LSXM.K stock at the discount to NAV, resulting in a closing of the discount in a tax efficient manner. Second, LSXM.K’s current debt capacity will allow for ongoing capital return. Even before consolidation, LSXM.K can still use leverage to buy back shares, given that the company only utilized $250mm out of the $1bn capacity of the margin loan. The additional debt capacity from this existing debt alone can be used to eliminate ~6% shares of LSXM.K at current price. Third, SIRI could enter into a Reverse Morris Trust transaction, in which SIRI issues equity to purchase LSXM.K. Given that LSXM.K would own 50+% of the pro-forma business, the transaction will be tax-free and SIRI will still be able to utilize the NOLs, but this essentially puts Liberty in control of SIRI’s cash flow without having to hit the consolidation threshold and effectively closes the discount.

Bankruptcy filings of terrestrial competitors will drive users to satellite radio. iHeartMedia and Cumulus Media, two of the nation’s largest terrestrial radio station owners, are currently in distress and may declare bankruptcy before YE2017. These companies together have 132mm listeners, and any disruption to their operations is likely to drive users to satellite radio, the closest alternative to terrestrial radio.

Rollout of next-generation 360L radio platform in 1Q18 will increase clarity around SIRI’s central role in connected vehicles. Software for 360L, which has the ability to integrate 4G and LTE into the car entertainment system, is scheduled for Sept. 2017 completion. 360L incorporates functionality for both android and iOS phones and enriches the SIRI content experience by also allowing customized music. Additionally, it is an extra avenue for data capture directly at the subscriber level. We are optimistic about SIRI’s connected vehicles business, which represents entirely unmodeled upside potential due to the lack of granularity at this early stage. We believe SIRI enjoys a clear first-mover advantage in the space, where they have been encouraged by OEMs to create a connected vehicles platform - indicative of their strong incentive alignment. This is reflected in their clear superiority in the reach of their partnerships with OEMs, which include Toyota, Hyundai, Audi, Fiat Chrysler, among others - these partnered OEMs comprise 58.4% of Total 2016 U.S. New Car Sales. This is in stark contrast to their competitors in Verizon (5.5%) and GM (17.3%), the latter of which is only pursuing a platform for itself. In addition, SIRI has already partnered with AT&T to provide 4G and LTE connectivity. While SIRI can make their radio an even more central component of future connected vehicles, OEMs can improve vehicle reliability through integrated roadside assist, remote vehicle locking/control, and vehicle condition monitoring which could preemptively alert owners of car issues before a breakdown. We believe that this segment represents a growth opportunity both in itself and to SIRI’s core business, and the rollout of 360L should provide more detailed visibility into this fact.

Valuation

We constructed four operating models to encompass our bear, base, base with levered buybacks, and bull scenarios. For the SIRI revenue build, we derive the potential subscriber TAM from a top-down US vehicle market model incorporating used/new vehicle penetration rates, conversion rates, churn, and car obsolescence rates. Compared to the sell-side, our DCF assumptions differ the most in our view on used vehicle penetration and churn rates, which are driven by our thesis points on them. We built operating leverage into our cost structure, with a 70% flow-through margin in line with Street estimates. Neither our base case nor bear case incorporates levered buybacks, nor spectrum value since the base case is subject to a 40% probability discount from our projected NPV. Our bear case represents a worst-case scenario, with growth dropping off quickly and costs continuing to increase. Our valuation was constructed with an Unlevered FCF multiple on our 2021E number; we believe this is more accurate than EBITDA given the company’s high cash conversion and stable cash flows. Moreover, this approach is not significantly different from a DCF approach, since we assume that before the terminal period SIRI will continue to use most of its UFCF to buy back shares. In our base case, applying a cautious 16x multiple (implying a 6.3% yield) on our above-Street UFCF number, we see an asymmetric 19.2% upside in SIRI, which translates into a 44.9% upside with a position constructed via LSXM.K. Our base levered buyback model, which we believe is warranted by management statements, adds an incremental 5.3% upside to SIRI and 6.5% upside to LSXM.K.’s base case upside.

Relevant Comps

For comps, the focus will be on the underlying SIRI stock, as LSXM.K is simply a tracking stock. There are no direct competitors in the satellite radio space due to the industry structure after the consolidation of Sirius and XM in 2008.

Some investors perceive terrestrial radio (AM/FM) players as comps since they compete for listening hours, but we do not believe that this is appropriate. First, AM/FM radio firms have a fundamentally different revenue structure, which is solely based on advertising, while SIRI generates revenue solely from subscription fees. Second, main terrestrial players are not currently suited to compare well to SIRI. CBS Radio is part of CBS, a diversified company - as a result, its metrics do not provide helpful context. Cumulus Media (CMLS) and iHeartRadio (IHRT) are in distress right now. However, we do note that pre-distress CMLS used to trade at around 12.1x EBITDA and 19.1x Unlevered FCF and IHRT used to trade at around 11.8x EBITDA and 18.1x Unlevered FCF.

Another perception of the market is that streaming players such as Spotify (private) and Pandora (P) are close competitors. We do not believe this comparison is appropriate either - these firms have a vastly different revenue structure, as they receive a considerable portion of their revenues through advertising and charge significantly lower ARPUs compared to SIRI. Additionally, they target fundamentally different customer bases - SIRI targets higher income, older consumers who own vehicles, while Pandora and Spotify attract younger and less affluent customers. Despite our strong belief that streaming players are not great comparable firms, we still examined Pandora’s historical trading multiple. We note that Pandora has never generated any positive EBITDA or cash flow, and has historically been trading at 2.1x revenue.

Further, while Charter Communications (CHTR) and Comcast Corporation (CMCSA) are not perfect competitors since they are much more asset-intensive and have less stable subscriber bases than SIRI, we also examined metrics on these companies as they are similar monopoly communication providers and therefore their metrics may offer some helpful context. We note that Charter and Comcast are currently trading at 18.6x and 23.5x Unlevered FCF respectively.

(Editors' Note: This is a republication of an entry in the Sohn Investment Idea Contest. All figures are current as of the entry's submission - the contest deadline was April 26, 2017).

Analyst’s Disclosure: I am/we are long LSXMK. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.