SPBU valuation analysis
We have seen posts claiming that SPBU can generate .05 EPS (for now let's not worry about when, as that will ultimately be a function of when they get to certain subscriber levels).
Valuation of SPBU requires making assumptions about a couple key parameters (especially revenue/subscriber/month) and then determining how many subscribers are required to get a particular EPS.
Here is a simple analysis, followed by a more incisive one.
Simple analysis.
EPS = Earnings / Share count, where Earnings = Revenues * net margin
therefore EPS = (Revenues * Net Margin) / Share count
How do we come up with the required numbers?
Revenue will be a function of subscriber count. I have seen estimates of $.20 revenue/subscriber/month. Is this reasonable? I don't know. That would be 20% to SPBU if the subscriber paid the carrier or insurance provider $1.00 per month for the service. These figures seem a reasonable place to start. $.20/subscriber per month equates to $2.40/subscriber/year in revenue.
Ameritrade shows 314MM shares outstanding for SPBU right now. I think there will be some more shares issued before they go cash flow positive (which the CEO has told us should come late Q3 or in Q4), so I am going to use 325MM shares fully diluted.
To get EPS = .05 on 325MM shares, that means that total earnings need to be $16.25MM.
What will net margin be? To make the numbers easy let's do it two ways - at 5% net margin and at 50% net margin. 5% net margin sounds low for a business that does not have significant cost of goods sold, and 50% may be high (though not out of the question for this niche).
At 5% net margin, revenues need to be $325MM to generate $16.25MM in earnings. At $.20/subscriber/month, that would mean 135MM subscribers are needed. Not happening. On the other hand, at 50% net profit, only 13.5MM subscribers would be required to generate .05 EPS given the assumed share count and revenue/subscriber/month. Very possible.
Obviously, simple math can be used to run through sensitivity cases in this simple model. The equation is linear against any of the variables. Increase anything "good" by 50% (i.e. if revenue/subscriber/month rises to .30) and the required subscriber counts at any net profit goes down by 33% (you would need 9MM subscribers at 50% net). Decrease anything good by 50% and the required number of subscribers doubles (at .10 revenue/subscriber/month, you need 27MM subscribers per month at 50% net or 270MM subscribers at 5% net to get to .05 EPS on 325MM shares. You can work through the math for any combination of sensitivity cases (i.e. a different share count).
What kind of subscriber count is feasible? I think 135MM or 270MM is out of the question until SPBU has many times the number of distribution agreements (where SPBU provides "white-label service behind the scenes) than we see now or see on the horizon. But counts of 5-10MM seem very doable, and counts of 10-30MM within the realm of possibility, given some basic parameters. Subscriber count = total number of cell customers to whom SPBU's service is made available via one of SBU's partners times the % of those that sign up for SPBU's service (adoption rate). The table below indicates how this would work.
Customers for SPBU carriers -- Adoption rate -- SPBU subscribers
50MM -- 10% -- 5MM
50MM -- 20% -- 10MM
100MM -- 10% -- 10MM
100MM -- 20% -- 20MM
150MM -- 10% -- 15MM
150MM -- 20% -- 30MM
The problem with the simple analysis is the big swing tied to net profit %, which itself is probably highly sensitive to subscriber count. So, instead of swagging a net profit percentage, let's identify the revenue (and therefore customer count) required to cover fixed costs and then generate a given amount of earnings.
In this more sophisticated model:
--the first n customers generate revenue which will cover the fixed costs
--the revenue from each subsequent subscriber is split between covering variable costs associated with that subscriber and falling to the bottom line as net income (earnings)
What are fixed costs? Given that SPBU is transitioning and rolling out service, I think the most recent quarter is a better place to start than the trailing 12 month figures. In the March 2012 quarter, SPBU had SG&A expenses of about $1.6MM, including about $265K in R&D expense. That annualizes to $6.4MM. The CEO has indicated they are making significant inroads on the monthly G&A, and I imagine R&D will drop now that they are rolling their services out, so to have convenient numbers I am going to use $6MM as the annualized fixed cost figure. I actually think that may be on the high side, but let's stay conservative.
Our revenue assumption is that each subscriber brings in $2.40 per year in revenue. $6MM in fixed cost divided by $2.4MM revenue per subscriber per year gives 2.5 subscribers required to cover fixed costs. (I know that there are actually some variable costs associated with these 2.5MM subscribers, but for simplicity I will ignore that as I think the other assumptions are collectively on the conservative side). If you think the fixed cost number is low, have at it. Every extra $1.2MM in fixed cost requires another 1MM subscribers to cover it.
As each new subscriber is added, they bring in an additional $.20 per month in revenue. How much of that falls to the bottom line, and how much goes to pay for more (fully automated) backups, some additional storage space and servers, and some technical support and customer support calls at a help desk? This is a very scalable business model, so it is unlikely that on-the-margin costs are anywhere near on-the-margin revenue as subscribers are added. If we very conservatively assume that a full 50% of the extra revenue has to be used to cover on-the-margin costs (this seems REAL high to me, my gut says it is closer to 20% than 50%), then each new subscriber after the initial 2.5MM drops $1.20 per year directly to the bottom line. Simple math tells us that 13.5MM (ish) additional subscribers generate the requisite $16.25MM in earnings needed for .05 per share. This means a total of 16MM (ish) subscribers, for a full year, gets us to .05 EPS after covering all costs.
This number does not seem out of the question at all. Maybe not in 2012, but I can certainly see that kind of number 12 months out from here.
At this point it looks like getting to .05 EPS is feasible, but to me doing so in 2012 seems a big stretch. However, I think that is the wrong thing to focus on. Who really cares about .05 EPS anyway? To get to .05 EPS, SPBU would first have to get to .01 EPS, and then .02 EPS, etc. If SPBU were earning .01 EPS, what kind of PE would it get? For simplicity, let's say a PE of 12, which would yield a price of .12--more than a triple from today's price. Good enough for me to look at SPBU as a potential multi-bagger.
To get to .01 EPS, using the fixed and variable cost model, you need FAR FEWER subscribers. You still need 2.5MM to cover the fixed costs, but you only need 2.7MM more to get that .01 EPS keeping all the other parameters the same. 5.2MM subscribers seems more than do-able, in fact, very likely.
With this analysis, 5.2MM subscribers gets us to .01 EPS and a price of .12 (a triple from here) if the market were giving SPBU a PE of 12 at the time. However, SPBU would almost certainly be growing revenues and earnings rapidly at that point. For a PE of 12 to be "fully valued", EPS growth rate would have to be 12% per year. It is almost mathematically impossible for SPBU to ramp from where it is to .01 EPS and have only a 12% growth rate at that endpoint (the subscription count would need to ramp quickly to 5.2MM and then plateau, which is really unlikely!). If SPBU were growing earnings at 25% per year at that point (still very hard to do mathematically) then the PEG would be .50 if the price were .12, and would suggest that the PE should expand toward 25 (and therefore a price of .25).
In summary, it seems like a very do-able subscriber count (5.2MM) can get SPBU to a point where it could triple and still be very undervalued. Rather than worry about whether it can triple or be a five or ten bagger or more, I'd rather stay with conservative assumptions and say the analysis is good enough for me to risk some speculative capital at this point. Hopefully, we can see some actual data that allows us to fine tune some of the parameters, and will see some actual subscriber count data going forward.
