Thursday, March 29, 2007

Some Comments on The Criterion Report

March 29, 2007

Comments on the study conducted by Criterion Economics against the merger.

SSG is looking over the study conducted by Criterion Economics and commenting on various aspects of the study. Bear in mind that we are not professional economists, but we can offer insight and opinion that we feel the author of the report has overlooked. This analysys will happen over a series of articles.

The author of the report states that in mid 2005 XM raised their price but did not suffer an increase in churn, and therefore there is a distinct difference in SDARS and other providers of audio content. His argument is that the price increase did not cause consumers to seek out new providers for content.

SSG Comments:

First, churn needs to be defined and the other impacts happening in that time period need to be considered.

1. Churn is a measure of cancellations relative to the average size of the base. An influx of subscribers above the norm can have a direct impact on the reported churn rate.

A. General Motors introduced their highly successful employee discount program and had record sales. This created an inflation ion the OEM channel subscriber numbers for Q2 and Q3. There was then a lower than expected OEM number in Q4, but the retail holiday season helped offset that issue. The author of the report fails to acknowledge this fact.

B. During that same timeframe, XM satellite radio launched an aggressive friends and family promotional program. During this timeframe the friends and family program increased from about 12% of the subscriber base to 22% of the subscriber base. Consumers were enticed with free hardware offerings and/or free service for a period of time. The author of the report fails to acknowledge this fact.

2. During this timeframe XM Satellite Radio saw an increase in Subscriber Acquisition Costs (SAC) and reported losses greater than anticipated. The author of the report fails to acknowledge this fact.

3. XM dropped a premium tier of programming and made it part of the base service. Thus, there was an increase in value and level of service that the consumer received. The author fails to acknowledge this in his report.

4. XM had elasticity in price because Sirius was already at $12.95 per month. This gave XM room to increase the price without a large backlash from their consumers. The author failed to acknowledge this in his report.

5. XM satellite radio also offered consumers the opportunity to lock in the current rate by pre-paying. Many consumers took advantage of this offer. The financial reports show that the average pre-pay subscriber increased during this timeframe. The author of the report fails to acknowledge this in his report.

Simple mathematics dictates that a surge in subscribers would increase the subscriber pool such that the normal cancellations would appear to be in line with a normal situation. The author of the report, and in fact most analysts that covered the sector, did not take into account the substantial changes and dynamics that happened in the XM subscriber base during that timeframe. There was a substantial increase in OEM (estimated at 175,000 to 225,000 higher than normal over 2 quarters) subscribers as well as a substantial increase in Family plan subscribers. The author speaks of a price increase, but does not consider a few hundred thousand units that were selling with three months of free service, and $6.99 per month thereafter (family plan). For the author to state that churn was not affected without getting to an apples to apples comparison is simply wrong.

The increase in SAC speaks to what was happening. Simply stated, XM was in the mode of “buying” subscribers at a rate higher than they had previously been spending. Thus, the price impact did indeed have an impact on metrics…..it simply was not seen in churn. Ask an XM investor about the SAC costs during the second half of 2005 and you will see a direct impact.

The change in price was coupled with programming being removed from a premium tier. Thus, the base package of XM was increased. This added consumer value. This coupled with the option to lock in pricing helps to offset churn.

Simply stated, on this aspect of the report, the author seems to have missed some key components and factors. Would these factors change the authors opinion or number on this section of his report. We may never know. The report as it exists was contracted by an anti-merger group, and I somehow do not see that group asking the author to address these factors, nor do I see the author changing his report to take these items into consideration unless he was paid to do so, or there was another compelling reason.