As most investors in Sirius XM Holdings (NASDAQ:SIRI) know, the company has one basic method for adding subscribers. It subsidizes the inclusion of its radios as standard equipment in about 70% of all new cars sold in the US. By offering the new car buyer a free trial, the expectation is that a certain percentage will decide that the commercial free music or other content offered by Sirius is sufficient to induce that buyer to becoming a self-pay subscriber. Nearly three quarters of the trial run by Sirius will come through this new car channel.
In addition, the company has a program that offers 90-day free trials to certain buyers of used cars, although the expectation is that fewer of those buyers will become self-pay subscribers (a number Sirius management has referred to as being in the low-30% range). Sirius also has a program that offers certain owners of cars with in-active radios a 60-day free trial when they bring their car into participating service centers for maintenance or repairs. All of these trial programs are dependent on the car having an OEM-installed satellite radio.
On the Q2 conference call, the company once again discussed the cost to subsidize the OEMs for the installation of the radios and its impact on earnings. CFO David Frear stated:
Adjusted EBITDA grew nearly 31% to a record $370 million. And adjusted EBITDA margin jumped 570 basis points from 30% to 35.7%.
Similar to the first quarter, the biggest single factor was the drop in subscriber acquisition cost down 18% as the growth in the OEM installations was more than offset by a drop in unit costs. We pick up 4.1 points of margins on the improvement in SAAC [sic] alone.
An 18% drop sounds impressive, doesn't it? The only problem is, I can't find the numbers to make this 18% figure work. The line item for subscriber acquisition cost from the second quarter 10Q P&L shows (only selected shown):
Results of Operations
Set forth below are our results of operations for the three and six months ended June 30, 2014 compared with the three and six months ended June 30, 2013 .
2014 vs 2013 Change
2014 vs 2013 Change
For the Three Months Ended June 30,
For the Six Months Ended June 30,
Cost of services:
Revenue share and royalties
Programming and content
Customer service and billing
Satellite and transmission
Cost of equipment
Subscriber acquisition costs
The above shows subscriber acquisition costs declined 4% from the year ago quarter. And, that's consistent with the decline discussed in the 10Q that states:
For the three months ended June 30, 2014 and 2013, subscriber acquisition costs were $124,407 and $129,992, respectively, a decrease of 4%, or $5,585, and decreased as a percentage of total revenue.
What about the first quarter? Those costs were $123,022,000, and since that's an increase, Frear can't be referring to a sequential decrease. What about the metric SAC per gross subscriber addition?
These costs were reported as $33 for Q2 2014 and $47 for Q2 2013, a decline of $14, or 30% from the prior year. For Q1 of 2014, the SAC was $35, so on a sequential basis the decline was $2 or 6% from the prior quarter.
Subscriber acquisition costs remain a very large expense for the company, even though the subsidy per gross new car installation has declined significantly. If one looks carefully at the 10Q one can discover the reason for parts of the sharp drop in SAC per gross subscriber addition. The following is from a note in the 10Q about the change in certain purchase price accounting adjustments that affected adjusted EBITDA and SAC:
The purchase price accounting adjustments related to the Merger, include the: (I) elimination of deferred revenue ..., (ii) recognition of deferred subscriber ... and (III) elimination of the benefit of deferred credits on executory contracts, which are primarily attributable to third party arrangements with an OEM and programming providers. The deferred credits on executory contracts attributable to third party arrangements with an OEM included in revenue share and royalties, subscriber acquisition costs, and sales and marketing concluded with the expiration of the acquired contract during 2013.
The expiration of that acquired OEM contract during 2013 was the major reason that the SAC figure declined from the $47 in Q1 and Q2 of 2013 to the $33-$35 for the first half of 2014. Those adjustments added more than $22 million to SAC each quarter in the first half of 2013, or nearly $8 per vehicle. In addition, the increased dollar margin on the direct sales of parts and accessories in 2014 relative to 2013 ($31.8 million vs. $24.1 million) reduced SAC by more than $1 in the first half of 2014. Without these two changes the subsidies to OEMs would have decreased by $4.38 from $43.30 to $38.92 per vehicle, or about 10%.
While the decrease is impressive, investors should be aware that the subsidy model is still very costly for Sirius XM. The year to date total puts the cost of these subsidies on track to again approach half a billion dollars in 2014. And, since only 42% of the new car buyers will convert to becoming self-pay subscribers, more than half of that cost is "wasted". Unfortunately, there is no way of knowing ahead of time which of the 42% will convert.
A Bit of History
A combination of rising vehicle production and increasing OEM penetration rates have been driving total reported subscriber acquisition costs higher for the past several years. Since 2011 the costs have risen from $434.5 million to $474.7 million in 2012 and $495.6 million in 2013. This year may finally see a decrease in this expense as the year to date figure is only 1% ahead of 2013 despite a 12% increase in installations, and the most recent quarter is showing a 4% decline in total expense despite a 10% increase in installations from last year's second quarter.
It should also be noted that each incremental self-pay subscriber is quite expensive to acquire, and that expense has been growing recently despite an influx of used car net additions. In 2011 when the company spent $434.5 million, it added 1,221,943 self-pay subscribers at an average cost of $356. In 2012 when the self-pay net adds climbed to 1,661,532 and the cost rose to $474.7 million, the cost of each of those incremental net adds fell to $286. Since then, the costs have been rising.
As the total cost rose to $495.6 million in 2013 and the self-pay net adds declined to 1,511,543, the cost per incremental self-pay net add rose to $328. So far this year, with the costs slightly ahead of last year's pace and the self-pay net adds significantly behind - only 553,191 vs. 727,462 in first half of 2013 - the cost for each of those incremental self-pay subs is $447. Assuming the company is able to achieve its target of 1.25 million net adds for the year, and the total expense comes in at $487 million (slightly less than the current run rate), that cost could decline to $390 each, but that would still be the highest in 4 years.
I have been unable to determine how Frear calculated there was a "drop in subscriber acquisition cost down 18%", although it is clear that the subsidy per new vehicle has declined. It should also be recognized that the cost to acquire each incremental self-pay subscriber continues to rise.
The combination of rising new vehicle sales and increased penetration rates has been offset by lower conversion rates and cancellations from a large installed base. Even the addition of a significant number of used car re-activations (which incur no subscriber acquisition costs) has been unable to slow the increasing cost of adding each new self-pay subscriber.
It will be interesting to see whether the SAC metric that has fallen sharply this year can continue to trend significantly lower and whether self-pay subscribers will begin to pick up. These are important metrics for investors to follow.
Disclosure: The author is long SIRI. The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it (other than from Seeking Alpha). The author has no business relationship with any company whose stock is mentioned in this article.
Additional disclosure: In addition to my long positions, I have January 2015 $4 covered calls written against portions of my long positions in Sirius XM. I also trade blocks of Sirius XM on a regular basis.