Description

It is hard to describe this idea as a “value” trade but I believe that an investment in the bonds of XM Satellite Radio (XM or the Company) is a compelling speculative investment. XM is one of two providers of satellite digital radio service in the United States. XM rolled out its product nationwide in November 2001 and has exceeded most analyst expectations for new subscriber additions. Its problem is that the capital markets have turned against negative cash flow projects and the Company is rapidly running out of cash. XM needs $500 to $600 million to finance its business plan, but only has enough cash to last until the early months of 2003. The stock of XM is extremely speculative, but a purchase of the 14% Senior Secured Notes due 2009 (the 14% Notes) will either generate a bond return of 4x over the next 3 years or will be converted into common equity at what is likely to be an attractive valuation.

The Opportunity

XM and Sirius Satellite Radio are competing to provide digital quality, mostly commercial free radio to the American driving public. I won’t bore you with the technological details, but I will say that I have an XM radio and that it was relatively easy to have installed (I live in Manhattan and nothing is easy) and that the product is great. The sound quality is excellent, the programming is good (even if there is way too much Paul McCartney, Yes and Fleetwood Mac) and the reliability is good (If you are driving in very hilly terrain with overhanging trees, the signal can be erratic. Otherwise, the only service interruptions are from some overpasses, although not all).

The real payoff for XM is not going to come from aftermarket sales (which is XM’s only channel to date), but in the OEM market. General Motors, a significant investor in XM, has announced that XM will be available as a factory installed option on 25 GM 2003 vehicle models. This is a fairly aggressive rollout, which is likely to accelerate for model years 2004 and 2005. The satellite radio option will cost $325 (which GM expects to reduce to $200 for model year 2004) and the service costs $9.99 per month.

XM has performed all kinds of marketing studies, which indicate that the potential market is huge, perhaps as large as 50 million by 2012. But let’s not get ahead of ourselves, XM has fewer than 200,000 subscribers currently and should have 350,000 or so by the end of the year. Since the discrepancy between the current and potential market sizes is so large, I believe it is better to analyze the product from the perspective of the consumer (i.e. does it offer good value to the consumer) and that of XM (i.e. does it generate an attractive rate of return). If it seems like an excellent value to the consumer and generates good returns to XM (which will be protected from aggressive competition by its FCC license), then, at the right price, it should be a good investment. First, let’s examine it from the consumer’s perspective. Let’s assume that the only value to the consumer is the avoidance of commercials (I am ignoring sound quality and programming, which, I believe, are at least equally important, but are difficult to quantify. One hundred channels permits XM to fill a number of programming niches that cannot be filled even in the largest metropolitan markets). According to Arbitron, in 2000, motorists listened to the radio an average of 50 minutes a day, despite the fact that 92% of cars had a CD or cassette player. Let’s assume that an average subscriber is a more intensive radio listener and listens to the radio 125% of the average or approximately one hour a day. Let’s further assume that an average XM subscriber will listen to XM radio 75% of the time and local radio 25%. That means that an average XM subscriber will listen to XM radio 45 minutes a day (I am making up these numbers, but they seem reasonable). If an hour of local radio programming contains 15 minutes of commercials, then an average XM subscriber will avoid 11 minutes of commercials a day by listening to XM radio. If we assume that an XM radio costs $200 and the service costs $9.99/month, then the daily cost of XM radio (assume a 4 year radio life) is $.47 or about four cents per minute of avoided commercials. I don’t know about you, but all other things being equal, I would gladly pay four cents to avoid being urged to repair my muffler at Midas.

What about XM’s economics? Let’s assume that, over the long term, total SAC will be about $100, variable cost operating margins will be 70% and churn will be 1.5% per month (I am ignoring system operating costs, G&A, and satellite replacement costs because they are relatively fixed. If the marginal economics are powerful enough, then we can determine the necessary total subscribers needed to generate positive returns for the entire business. I am also ignoring advertising, because I am not convinced that ads will always be a part of XM’s business model). If we assume that GM (or some other OEM) takes 15% of the monthly revenue, then the after-tax IRR of a marginal OEM customer is 39%. Pretty good.

Now, how many customers will it take to defray system operating costs, G&A and satellite replacement costs, which, Morgan Stanley estimates at $50mm, $30mm and $100mm per annum, and generate an attractive return on XM’s investment? Coming up with a growth ramp for satellite radio and reasonable estimates for its ultimate market size is not easy. Will satellite radio be like TIVO, a product its users swear by, but which refuses to take off, or like satellite TV, which experienced extraordinary take-up rates? Arguments can be made both ways. I believe that the OEM dealer network will be a much more powerful distribution channel for satellite radio than the 10 or so square feet of Circuit City shelf space that TIVO commands and that satellite radio will fairly quickly become ubiquitous. Let’s assume, however, that the total market size is 40% of XM’s estimate or 20 million subscribers, which are split 50/50 between XM and Sirius. In that case, the income at XM will be:

Gross Revenues $1,200mm
OEM Payments (est.) (180)
Programming & Customer Costs (360)
System, G&A & Satellite Replacement (180)
Subscriber Acquisition Costs (180)
Operating Income $ 301mm

If we assume that such a business is worth 10x EBIT, then XM’s future value will be $3.0 billion, given these assumptions. Admittedly, 20 million satellite radio customers is a lot of customers, but there is an addressable market of 208 million registered private motor vehicles.

Is the price right? There are $350 million of the 14% Senior Secured Notes. The Notes have the next 2 coupon payments (payable 9/15/02 and 3/15/03) in escrow. That means that at the current price of 33, the stripped price being paid for the bonds is approximately 26 or a total valuation of XM of $90 million (there are $125mm converts, but they are subordinated to the 14% Notes). In order to execute its business plan, XM needs to raise (Morgan Stanley estimate) $525 million or $350 million if all of the debt of XM is equitized. Given the potential of the business model, it seems likely that this money will be raised. The question is how dilutive will it be to the existing investors at XM? If the money comes in junior to the 14% Notes (unlikely), then everything is for the best in the best of all possible worlds. XM will have 3.5 years to execute its business plan and the holders of the 14% Notes will receive 49 in coupon payments on an investment of 33. If the business plan works, the 14% Notes will be called at 107 in March of 2005. If it doesn’t, there will be another restructuring. In either case, bondholders will earn an excellent return. A more likely outcome is that the new equity investor (who is likely to be one of the existing equity investors) will demand that the 14% Notes be equitized or, at the very least, that the coupons for the next three years be converted to equity. Even if the 14% Notes are completely equitized and we assume $10 million of leakage to the converts and the existing equity, then so long as the $350 million of new equity ends up owning no more than 78% of the restructured XM, a buyer of the 14% Notes is purchasing equity in the restructured XM at less than the price paid by the new equity investors. That is a bet that I am willing to make.

The prices of XM common and the 14% Notes are quite divergent (there is no borrow available on the stock). If we assume that new equity will require 78% of the equity to finance the business plan (i.e. the current price of the 14% Notes), then the 14% Notes reflect a post-new money valuation of XM of $451 million. The current market price of XM’s common reflects a post-new money valuation of almost $1.5 billion. Thus, in order for the stock to outperform the 14% Notes (assuming the new money is invested at the current stock price), the 2005 XM enterprise value must be approximately $4.5 billion AND the new equity must be willing to invest $525 million in an unrestructured XM. Not bloody likely.

Catalyst

XM is running out of money. New capital will either be injected junior to the bonds, which will cause the bonds to rally, or the bonds will be converted to equity at an attractive valuation.