Last week, Sirius XM Holdings (NASDAQ:SIRI) CFO, David Frear presented at the Bank of America Merrill Lynch Global Telecom and Media Conference in London. Following that presentation, a couple of comments on my previous article pointed out that Frear said growing ARPU wasn't of great importance. The quote:
And I [get asked] about ARPU a lot and to be honest, it's not a measure I spend a lot of time focusing on. I do focus on the price structure of the business, but when you think about it, with 80% of households having two or more cars, if we were really successful, ARPU would drop, because we offer a multi-radio discount, which is below our current ARPU.
So we would love to end up with lower ARPU, but a lot more subscriptions in the long run.
The above quote was part of a longer response to a question by Jessica Reif Cohen, the Merrill Lynch analyst interviewing Frear. She had asked:
80% of cars, car-owning households in the U.S. have two or more cars. Do you have any idea [of Sirius's] penetration within those secondary and tertiary car households where you already have at least one established relationship?
I would like to point out that Frear didn't directly answer the question. The rest of his answer to that question reminded me of a politician where the answer went off on a tangent to avoid answering the question. Here is the complete answer:
Let's see. I've got it a little bit differently than I think [the way you] asked the question, a little over 20% of our subscribers are multi radio households. And therein I think lies the great opportunity because as you said, 80% of households have two or more cars.
Now only 20% of households have new cars only, so do not only the 80% have two or more cars, 80% have used cars in the household. So I think we have a great opportunity to take that 20% number on multi radio households towards the 80% over time.
And I [get asked] about ARPU...
While this is the first time I have heard the explicit comment that "we would love to end up with lower ARPU, but a lot more subscriptions in the long run", it is not the first time that the company has discussed using marketing tools that sacrifice ARPU in order to increase subscribers. In the past, it was typically part of the discussion on offering retention discounts. The more interesting question is why didn't Frear discuss the number of households with multiple subscriptions? (To be clear, it should be noted that the number of subscribers is actually the number of subscriptions, and a household with three subscriptions, even if all are billed to the same subscriber, is considered three subscribers.)
This is important for several reasons, some generally positive and others generally negative. On the positive side, many Sirius subscribers are extremely loyal. The fact that some may take more than one subscription probably increases the loyalty and stickiness of that customer to the service. Furthermore, if fewer households are actually subscribers, that means that there could be many more households out there that could potentially become new subscribers. On the flip side, if the 21.3 million self-pay subscribers are already concentrated in fewer households, it could indicate that a disproportionately large share of households have already rejected the idea of paying for radio after having had a free trial. It could also indicate that the incremental penetration into multi-car families where there is already one subscription has already been achieved.
I have been hearing about multi-subscription customers for as long as I have been following Sirius on Seeking Alpha. There has been a fairly constant stream of claims similar to "I have a subscription for each of my cars", "I have one in my car and I had to get one for my wife's car", or "I got one for my girlfriend's car." There have also been some complaints from some that they should get a bigger discount for second and third subscriptions.
This is nothing new to the company. I wrote about it in an article in October of 2012 titled "Sirius XM CEO Discusses Household Penetration Rates", based on a presentation that then-CEO Mel Karmazin made at one of the annual Liberty Media (NASDAQ:LMCA) investor days. Here is an excerpt from that article, beginning with the quote from Karmazin:
"... the important reason I put this slide up is for you to take a look at the percentage of households that are paying for radio. This isn't the number of subscriptions, because people have multiple subscriptions, but today, only - and I say only - 13% of the 110 million households in the United States have said that they want to pay for radio. We believe that number could be a whole lot bigger.
... We think this number is low... We can't give you, today, how far it's going to go."
He continued by making a comparison to other pay services and said that Sirius XM penetration is not likely to go to the 90% penetration of pay television. Clearly, there is room to penetrate more households - a lot of room. But what else do the numbers tell us? 13% of 110 million households tells us that Sirius is in 14.3 million households. It also tells us that the 23,365,383 [total paid] subscribers are spread over those 14.3 million households, and that each household has an average of 1.63 subscriptions.
It would have been great if Frear gave an update on household penetration rates to see how much, if any, progress has been made on increasing penetration above 13%. Since Karmazin presented those figures, total subscribers have risen 10.5% to 25,826,109, and self-pay subscribers have risen 11.6% from 19,041,519 to 21,255,297. Unfortunately, we don't know if the household penetration rate has increased, decreased or remained the same.
Regardless, here's the issue. If we assume that the company still has an average of 1.63 subscriptions per household, it already has a substantial number of subscriptions at the multi-radio discounted rate. One question investors should be asking is whether the ratio of multi-radio households is likely to climb much higher.
Here is another point to consider about whether or not multi-radio households would drive down ARPU. According to the company's web site, the standard monthly rate for a second radio is $9.99. In addition, the standard monthly music royalty fee (or MRF) is $1.25 per month. In addition to these two fees, the company also includes advertising revenue in the calculation of ARPU.
In the first quarter, the company had $22,214,000 of advertising revenue. Based on the weighted average number of subscribers outstanding during the quarter (25,602,139), the advertising revenue contributed $0.87 during Q1, or $0.29 per month. In other words, the total standard contributions for a second radio on a multi-radio family plan is $9.99 + $1.25 + $0.29 = $11.53.
Considering that ARPU was $12.18 during Q1, and the standard monthly rate for the first radio is $14.99 plus a $1.75 MRF and the same $0.29 for advertising, there is much more involved in bringing ARPU down than just the multi-radio discount. Part of this would be the company's discount for paying one or more years in advance. Another piece would be a presumed lower price paid by the OEMs for paid promotional subscriptions. (The company also charges premiums for certain extra packages, and offers other discounts for taking more limited offerings.) However, the biggest drag on ARPU is likely to be the company's extensive use of acquisition and retention discounts. At one time, Frear said these discounts were offered to "about 14% of the self-pay base," and nothing in the quarterly or annual reports since that statement was made indicates that there has been a decline in the use of that practice.
Aside from the multi-radio discount having a limited effect on ARPU, there were a couple of other points that are worth discussing. Frear stated "And I [get asked] about ARPU a lot and to be honest, it's not a measure I spend a lot of time focusing on." My immediate reaction was, why not? It has been listed as one of the key metrics for the company for a long time. It is a common measurement used by media and telecommunications companies to ascertain whether they are growing the revenue they get from each customer. It was also a number the company thought was worthwhile putting on a slide at its annual meeting when proudly proclaiming that it generated $149 per subscriber last year compared to Pandora (NYSE:P) at $8 and Clear Channel (NYSE:CCO) at $12.
Frear also discussed competition from terrestrial radio and companies like Pandora and Clear Channel's iHeart Radio. It is worth remembering that these are competitors to Sirius XM. They are part of the reason that new car conversion rates are at 42% and used car conversion rates are in the low-30% range. And competition is the reason that the company has to offer several forms of discounting.
It was disappointing that Frear chose not to answer the question about the household penetration rate, and it was puzzling that he would ascribe an increase in multi-radio households as a reason to expect ARPU to decline in the future. Growing revenue and subscribers is certainly important for Sirius, but so is growing revenue from each of those subscribers. And when Frear tells us that ARPU is not a measure he spends a lot of time focusing on, it could very well be a signal that it is becoming increasingly difficult to grow that number.
Disclosure: I am long SIRI. 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 have $3.50 and $4 January 2015 covered calls written against a portion of my SIRI positions, and will also frequently trade shares of Sirius.