Subscriber acquisition costs decline. Total subscriber acquisition costs were $124 million in the fourth quarter, or just 12% of adjusted revenue, the lowest percentage in the Company's history. The improvement in SAC was driven by lower subsidy rates per vehicle. SAC per gross addition was $44, a record low, a decline of 19% versus the fourth quarter of 2012.
And, during the conference call, CFO David Frear started addressing SAC less than a minute into his prepared remarks:
Q4 revenues were up 12% over the prior year as total subscribers grew to 25.6 million and self-pay subscribers grew to 21.1 million. A half a point increase in contribution margin to 69.8% set the stage for explosive growth in the EBITDA margin as SAC per gross add fell 19% to $44.
Adjusted EBITDA in Q4 grew 41% to a record $325.6 million and EBITDA margin expanded by almost 7 full percentage points as SAC fell by 4 points on a percentage of revenue basis and fixed cash operating expenses grew just 3.5% against the 12% revenue growth.
It all sends great, doesn't it? It would appear that the cost to acquire subscribers is falling rapidly. Or, is it? There is a line item in the P&L labeled "Subscriber acquisition cost". The definition from the 10K for the GAAP line item in the income statement is:
Subscriber Acquisition Costs include hardware subsidies paid to radio manufacturers, distributors and automakers, including subsidies paid to automakers who include a satellite radio and subscription to our service in the sale or lease price of a new vehicle; subsidies paid for chip sets and certain other components used in manufacturing radios; device royalties for certain radios and chip sets; commissions paid to automakers as incentives to purchase, install and activate satellite radios; product warranty obligations; freight; and provisions for inventory allowances attributable to inventory consumed in our OEM and retail distribution channels. The majority of subscriber acquisition costs are incurred and expensed in advance of, or concurrent with, acquiring a subscriber. Subscriber acquisition costs do not include advertising, marketing, loyalty payments to distributors and dealers of satellite radios or revenue share payments to automakers and retailers of satellite radios.
Quite a mouthful. That line item did decrease as a percent of revenue, and for the quarter was down 2% to $124.1 million from $126.7 million from 2012. However, for the full year, the cost was $474.7 million, up from $434.5 million in 2012. That's a $40.2 million, or 9.3%, increase.
What should be clear by now is that the metric known as SAC is not the same as the GAAP line item "Subscriber acquisition cost." Is one measure more meaningful than the other? Should an investor be concerned that the P&L expense increased by 9%? Much depends on how one is trying to evaluate the business of Sirius XM.
SAC, or Subscriber Acquisition per Gross Subscriber Addition, takes the total from the P&L, and subtracts the margin on the sale of radios and accessories and adds back the purchase price accounting adjustments from the merger of Sirius and XM. Simply dividing the line item from the P&L by the Gross Subscriber Additions for the period reveals that the numbers for 2012 compared to 2013 are surprisingly close. There was a slight decrease of $0.47 (or just under 1%) from $49.36 in 2012 to $48.89 in 2013. That annual decrease appears to have been driven mostly by the increase in Gross Subscriber Additions from the used car reactivations. As Frear has previously pointed out:
So certainly, the used car distribution doesn't add the subscriber acquisition costs, because we only do the radio subsidy once, all right. And that line, subscriber acquisition cost is really driven almost entirely by new car installations.
As the used car re-activations continue to grow as a part of the Sirius XM business, and drive the Gross Subscriber Additions higher, SAC should continue to decline.
However, in Q4 where the large decline in SAC was highlighted, the purchase price accounting adjustment was $0 in 2013 vs. $21.2 million in 2012. And the benefit from the margins on the sale of radios and accessories was a $7.5 million difference in favor of Q4 2013. These combined to reduce the numerator for SAC by nearly $29 million for Q4. Offsetting this benefit was a lower figure for Gross Subscriber additions for the quarter, which declined from 2,553,489 in 2012 to 2,409,804 in 2013.
From my perspective, SAC is not a particularly useful measurement. It gives one the misleading impression that it is a cost to acquire a subscriber. It's not. It ignores the important costs of "advertising, marketing, loyalty payments to distributors and dealers of satellite radios or revenue share payments to automakers and retailers of satellite radios" that are often associated with attracting and retaining customers.
Even more important than not including all the costs is the way those costs are spread out. A more useful measure would be the cost to acquire each net additional subscriber. In 2012 the company added 2,007,512 total subscribers, of which 1,661,532 were self-pay subscribers. In 2013, the company added 1,658,974 total subscribers, of which 1,511,543 were self-pay subscribers.
There is no denying that the metric SAC declined by 19% in Q4 2013 compared to Q4 2012. It also declined for the full year, although to a lesser extent. And, as a percent of revenue, the line item Subscriber Acquisition Cost also declined. In dollar terms, the line item Subscriber Acquisition Cost rose by 9% for 2013.
It is also obvious that Sirius XM spent more money to acquire fewer subscribers in 2013, even using the limited costs that the company chooses to recognize. And, if Sirius XM included all the costs it spent to grow the subscriber base, investors would have a more useful tool to measure the true cost of maintaining and growing the business.