These are the numbers Sirius XM reports each quarter and that give investors and analysts insight into how well the company performed in the most recent quarter and how well revenue, earnings and free cash flow could be expected to grow in the future. They include the self-pay monthly churn, new vehicle conversion rate, average revenue per user (or ARPU) as well as two other pieces of information that may or may not be addressed in the 10-Q -- an update on the used car program and the new vehicle penetration rate. Taken together, these pieces provide the information needed to understand revenue.
Investors now know that ARPU increased from $11.77 per month in the first quarter of 2012 to $12.05 in the first quarter of 2013. However, it also decreased sequentially from the $12.12 figure reported for the fourth quarter of 2012. This article will closely examine this particular metric.
The current 10Q notes the following about ARPU comparisons between the current quarter and the previous year:
For the three months ended March 31, 2013 and 2012, ARPU was $12.05 and $11.77, respectively. The increase was driven primarily by the full quarter impact of the increase in certain of our subscription rates beginning in January 2012, an increase in subscriptions to premium services, and an increase in the contribution of the U.S. Music Royalty Fee, partially offset by subscription discounts offered through customer acquisition and retention programs, lifetime plans that have reached full revenue recognition, and fewer days in the 2013 quarter.
The two phrases in bold are the first time I can recall seeing these particular references to lifetime subscriptions and/or the number of days in the quarter. In fact, looking at the last three years of 10K's covering the period from 2008 through 2012, no mention was made about the impact of lifetime plans. And, looking at previous first quarter 10Q's, no mention was made about the number of days in the quarter. For instance, the 2012 first quarter 10Q made no mention of the fact that it had one additional day compared to the first quarter of 2011. Have lifetime plans and one additional day in the quarter made such a dramatic difference in these figures? And, if the difference in the number of days is significant from a revenue perspective, shouldn't there have also been an impact from the number of days on the cost side?
I would have liked to see more information about how Sirius XM allocates prepaid subscriptions. This is the extremely large item on the balance sheet labeled deferred revenue. At the end of the first quarter the total was nearly $1.7 billion. There are FASB guidelines on revenue recognition, and how prepayments can not be recognized while there are future obligations on the provider and until such time as the service is actually delivered. CFO David Frear made this statement about the revenue during the recent conference call:
Revenue was up 12% in the quarter to a record $897 million paced by 9.2% subscriber growth and 2.4% growth in ARPU to $12.05. Before you all ask me about ARPU, remember this year's first quarter had one less day. I know it doesn't sound like much, but one less day means one less day's revenue, that's about 1.1% or $9 million of revenue and $0.13 of ARPU.
I suppose it is possible that a single day can make a difference in the quarterly revenue and ARPU depending on how the company is required to recognize revenue. However, there are several other factors that were expected to drive revenue and ARPU increases.
In the prior article cited above, I wrote the following, "ARPU is expected to show growth in the first quarter." This was based on comments by CFO David Frear in response to a question by Bryan Kraft of Evercore Partners on the previous conference call. Kraft asked:
So is it safe to say that you would expect the RPU to continue to increase in the first and second quarter as the price increase rolls through?
The ARPU will continue to pick up. Remember, it takes a long time for things to roll through, and so believe it or not, we're only just shaking off, now, the effects of the reduction in the music royalty fee from $1.98 to $1.42 that went through in December of 2010. That negatively impacted our ARPU numbers all year. In the first quarter, we'll pick that up, and so ARPU should grow.
Earlier on that call Frear had stated:
As a result of the [Copyright Royalty Board] decision, we increased US music royalty fee from $1.42 to $1.81 on our basic $14.49 package as of February 1st.
It isn't just the negative impact of the music royalty fee (or MRF) decline going away that should improve the ARPU; since that fee was also increased on February 1st, subscribers would have seen a price increase and ARPU should have risen even more. Meyer later reiterated that point:
I can tell you, and if you're a subscriber whose account renewed in February, for instance, you have been notified of a change in the MRF, and a passing along of that.
And, it isn't just the dual impact of the music royalty fee that should help ARPU. In the same way that it takes time for the decrease in the music royalty fee to "roll through," it also takes time for price increases to roll through. In the prior article I also pointed out that the basic monthly subscription rate increased from $12.95 to $14.49 in January of 2012. The positive impact of that 12% increase should still be rolling through the subscriber base and moving the ARPU higher.
There are other factors that are also impacting ARPU. Meyer addressed one when he responded to Kraft:
Also remember that the percent of subscribers who come in through paid trials also impacts that number, and as those grow, let's say differently, and different rates quarter to quarter, that can also have a factor on suppressing ARPU, obviously, as the money comes from the automaker instead of the end user.
The presumption is that the revenue from paid promotional trials is less than the standard monthly subscriber rates, or else it would have a positive impact on ARPU rather than be "a factor on suppressing ARPU." So, how has the mix of paid promotional subscribers and self-pay subscribers changed from the fourth quarter of 2012 to the first quarter of 2013?
Since Sirius XM does not provide detailed breakdowns on average subscribers during each quarter, a rough approximation will be made as follows. We know the number of "paid promotional subscribers" and the number of "self-pay subscribers" at the end of the third and fourth quarters of 2012, as well as the end of the first quarter of 2013. Using these data points, a simple average for each category can be determined as follows:
|Paid Promo Subs||Self-pay Subs||Total Subs|
|End of Q3 2012||4,323,864||19,041,519||23,365,383|
|End of Q4 2012||4,330,062||19,570,274||23,900,336|
|Average for Q4||4,326,963||19,305,897||23,632,860|
|End of Q1 2013||4,478,566||19,874,660||24,353,226|
|Average for Q1||4,404,314||19,722,467||24,126,781|
Comparing the average number of self-pay subscribers to the total subscribers in the fourth quarter of 2012 to the first quarter of 2013, we find that the percentage has increased from 81.69% to 81.75%. This would suggest that the change in mix would also increase the ARPU, albeit very slightly.
There were several factors at work during the first quarter that should have had a positive influence on the growth of ARPU:
- The phasing out of the negative impact of the MRF at the end of 2010 referenced by Frear
- The increase in the MRF on February 1st 2013
- The continued phasing in of the increase in the basic monthly subscription from $12.95 to $14.49
The change in the mix between self-pay subscribers and paid promotional subscribers would appear to be slightly positive, but consider it neutral for this discussion. On the negative side there are:
- One less day in the quarter referenced by Frear that would have resulted in a change of $0.13 and brought ARPU to $12.18 ($12.05 reported plus $0.13) vs. $12.12 in the fourth quarter of 2012
- Ending of full revenue recognition from lifetime subscription plans
- Retention and acquisition discounting
The decline of ARPU from $12.12 in the fourth quarter of 2012 to $12.05 in the first quarter of 2013 remains troubling. And, even if we accept Frear's comment and reset the figure to a six cent per month increase, the lack of growth from what should have been positive drivers suggests an increased use of discounting to maintain and grow the subscriber population.
Perhaps the most troubling aspect of the weak first quarter ARPU number was the comment on the fourth quarter conference call by Frear, when he answered Kraft:
In the first quarter, we'll pick that up, and so ARPU should grow.
ARPU didn't increase in the first quarter. Frear should have known about all the other factors that would be at work, including the one less day in the quarter, lifetime subscription plans nearing the end of their revenue recognition and the necessity for discounting. Is it reasonable to conclude that Sirius XM has limited pricing power despite being what some may consider a monopoly?
Whether the decline in ARPU is a fact to be concerned about or a fiction created by one less day in the quarter is something each investor will need to determine.
Additional disclosure: In addition to my long positions, I have January 2014 $3.50 covered calls written against many of my long positions in Sirius XM. I also trade blocks of Sirius XM on a regular basis.