Description

CSTR’s acquisition of ACMI, recent stock price weakness, conservative guidance, and potential catalysts merit a third writeup (I wrote the prior one). Including the acquisition, I estimate CSTR’s 2005 numbers as:

$95m EBITDA
$71m CFO (Cash Flow from Operations)
$11m Maintenance CapEx
$60m Sustainable Free Cash Flow (CFO – maintenance CapEx)
$110m Net debt (does not include the $43m NOL deferred tax asset)
25.3 million shares

$476m Market Cap
$586m Enterprise Value

6.2 EV/EBITDA (5.7 if EV reduced by NOL)
9.8 EV/sustainable FCF (9.1 if EV reduced by NOL)

It is possible that investors focus more on the EPS number and P/E, which at first glance leads one to believe that the stock price is not particularly cheap. However, EPS is misleading as it does not include the benefits of the NOLs, much CapEx is expensed (R&D), and depreciation is much higher than necessary (i.e. No coinstar machine has ever been retired yet they are depreciated over 5 years). Looking at EBITDA and sustainable FCF, CSTR’s stock price trades at a no-growth private market valuation. I contend that this is a modest growth, moderate moat business whose sustainable earnings is significantly higher than people realize, due to cost savings opportunities, expensed R&D, and the large number of immature coin counters in the machine base.

The coin counting business continues to be a natural monopoly business with nice returns in spite of the loss of Safeway and less profitable reworked contracts with the largest customers. Revenue growth is fairly predictable and has slightly exceeded my expectations according to the machine aging model I laid out in my prior write-up. I expect coin counting revenue per machine to continue to ramp according to this same model. For more detail about the coin counting business, see my prior writeup.

Coinstar’s stock price has fallen from over 27 to below 19 over the past few months. Some of this may be due to general market conditions, some may be due to investor concerns. I would guess that investor concerns are not related to the coin counting business, but rather with capital allocation and various short term factors. Specifically, was the ACMI acquisition worth while, and will all the CapEx spent on growth initiatives prove to be worthwhile? Does the lackluster Q1 forecast foreshadow years of pain to come from higher interest rates and higher gasoline prices? Below is my attempt to answer these questions.

ACMI ACQUISTION

Being a value investor, I wasn’t particularly thrilled to see CSTR acquire ACMI, a relatively low growth, low return business. On a positive note, ACMI is #1 in their niche which gives them scale advantage over the competition, and that only widens by combining with CSTR. On a per unit basis, ACMI’s skill cranes and vending machines take two months to ramp after installation then do not grow thereafter. So new growth requires CapEx to install new machines. At the time ACMI was purchased, CSTR could have purchased its own stock price at a similar EBITDA multiple and gotten (in my opinion) a superior business.

Management claims that the main reason for the acquisition was for cross selling opportunities (and, related to this, the ability to present a suite of services for the front end of retailers). My researches indicate that cross selling has worked within ACMI’s industry, and that has been one of the primary reasons for consolidation within that industry. The two sales forces have only recently been cross trained, so it is too early to assess whether cross selling will turn out to make the combination of these two companies better than 2 stand alone entities. Significant coin counter customer announcements such as Wal-Mart, Publix, or Tesco (in U.K.) would be a big deal for CSTR and would likely cause the stock price to rebound. I’d rather see new Coin counters installed with ACMI customers than the other way around, since coin counters are higher ROI investments.

In general, the company has taken a slow, cautious approach to integration, and has not touted the potential for synergies and cost savings on any press releases or conference calls. However, I see cost savings as a potentially huge benefit of this acquisition. A very large operating cost for both companies is the service technician network. If each technician could service 2/3 of the number of stores currently served (by being able to service equipment from either ACMI or coin counters), then there would be significant savings as each technician would spend less time traveling, and more time servicing. This would widen the cost advantage versus existing and potential competition. I spoke with CFO Brian Turner about this a few months ago and he indicated that they plan to cautiously test service network integration starting this summer. Each company has different incentive structures for the technicians, so the company wants to make sure that they do not hurt the culture and therefore technician effectiveness of either company. While some investors get excited about hyped up post-merger “synergies,” I actually prefer CSTR’s low key, cautious approach and conservative guidance.

GROWTH CapEx

In 2005, Coinstar expects to spend $11m on maintenance CapEx, $9m on skill cranes and vending units, $12.6m on coin counters, and around $20m on other growth CapEx for a total of $50m to $55m CapEx. This assumes NO cross selling occurs. According to CFO Brian Turner, the main reason for the secondary offering in December was because financial covenants of their debt agreements prevent CSTR from exceeding $60m CapEx per year. The lenders agreed to modify this covenant if CSTR would raise some equity and pay down a portion of the debt. In retrospect, CSTR’s timing was very good as they managed to price the secondary at $25/share when the market was hitting its peak. They did not use their stock as currency to acquire ACMI when it was lower 6 months earlier, a plus in my view (good capital markets discipline). I think they have a very good CFO in Brian Turner.

I have already expressed that I am not thrilled with the skill crane and vending businesses, but note that only $9m of CapEx goes towards these units. $12.6m goes to the coin counters, with 800-1000 installations expected. According to my model, coin counters earn high returns on investment, especially the ones placed in the U.K. So the more CapEx they spend on new coin counters, the better. I would be especially thrilled if they can get Tesco, the #1 grocer in U.K.

The other discretionary CapEx items break down as follows (unspecified amount to each):

Coins to Cards: $300 for a single card unit, considerably more for multi-card units. Last year, CSTR tested converting coin to Starbucks gift cards in several markets. This is a very exciting initiative as it allows customers to retain 100% of their coins in dollars as opposed to the usual 91.1% (Starbucks pays CSTR instead of the customer). This could tap into customers who were never willing to pay CSTR’s fee, and could generate more awareness of CSTR through cobranding. Both Coinstar and Starbucks were happy with the results and CSTR recently announced they would roll this out nationally, with several partners beyond Starbucks. However, the rollout may be delayed until CSTR finds an economic method for installing multi-card units in the machines.

ePay: $3500 for stand-alone units. Coinstar is rolling out ePay capabilities bundled with new Coinstar machines and as standalone units. They allow customers to top up phone cards and credit cards, and now have capabilities related to payroll cards. The company has not put out enough information for investors to draw their own conclusions as to ePay’s potential. Management has occasionally claimed that the goal is for the ROI on ePay to be similar to the coin counters, which would be great if true. For now, all these ePay investments depress earnings and management does not believe EPS will materially benefit until at least 2006. This is consistent with the ramp-up of the coinstar machines many years ago when R&D expenses and low density meant that CSTR was unprofitable for many years, despite the coin counters being good long-term investments.

Upgrades: a portion of CapEx goes to upgrading coinstar machines to be ePay enabled and to have fancier backlit screens. The company has found that backlit screens provide an immediate boost of 10% to sales.

So what to make of all this? I don’t know if all of CSTR’s growth initiatives will work out, but the coin counters are good investments, the vending and skill crane machines are probably not bad investments, and I think there’s a good chance that Coin-to-card turns out to be a very good investment as well. While this is the least certain part of the story for CSTR, it would take a pretty horrible ROI on the blended average of all these growth initiatives for CSTR to be a bad investment, at today’s stock price.

SHORT TERM ISSUES

1) CSTR recently issued Q1 guidance which fell significantly short of analyst expectations. CSTR has a history of issuing conservative guidance and nearly always performing near the top end of guidance, but the top end was below analyst forecasts. Management cites coin-to-card R&D and normal seasonality for the Q1 forecast. I don’t have a strong opinion about Q1 and don’t think one quarter’s performance matters all that much, particularly if it has to do with fluctuating R&D costs and Sarbanes-Oxley implementation.

2) CSTR took out a variable interest rate loan to acquire ACMI. Steadily rising short term interest rates will steadily increase their interest costs (thought interest rates are capped due to hedging). Though this will have a real impact on cash flow and EPS, the amount of debt is modest and CSTR could pay it down rapidly if they chose to slow down their growth initiatives. Even without the growth initiatives, CSTR generates substantial FCF each year that can be used to gradually pay down this debt.

3) Rising gasoline prices has a negative impact on operating expenses, given that services technicians drive large numbers of miles per year.

4) Sarbanes-Oxley expenses in 2004 and 2005 ($1.5m each year)

5) The vending machine portion of business suffered about a 10% sales drop starting last fall due to an industry-wide recall. These machines did not have recalled items but consumers nevertheless temporarily avoided the category. Management expects consumer memory to fade by the end of the year, and therefore sales rebounding to normal levels by Q4.

Catalyst

Short term issues dissipate with passage of time
Cost savings from combining the servicing networks from ACMI with CSTR
Getting a large new customer (Wal-mart, Publix, Tesco)
Getting Safeway back as a customer
One of CSTR’s new growth initiatives succeeds in a material way
CSTR acquired by a larger, related company