Those familiar with my work know that when it comes to growth stocks my main concern is about momentum in fundamentals continuing. Not only do I believe this to be the case with Equinix (EQIX), but I also think fundamentals and positive earnings surprises are now set up to potentially accelerate.
Equinix operates data centers. The company divides up space in these datacenters into cabinets. Customers effectively pay monthly for space and power. Customers also pay a monthly fee for each connection they make with other Equinix customers. The beauty of Equinix's business model lies in the fact that it has the leading position in almost all of its markets. When Equinix opens a new datacenter in a given market, it makes sure to tether that new building to all of its existing datacenters in that same market. Because customers prefer to be located close to other customers, they usually are willing to pay a premium to be in an Equinix data center versus the space offered at competitors. The significant bandwidth savings most customers get from being able to cheaply interconnect to other Equinix customers more than offsets the premium they pay. Furthermore, once a customer goes through the trouble of placing its servers, storage and Telco equipment in an Equinix data center, the costs and potential disruption involved in switching to a different provider is very high. All this gives Equinix a strong competitive advantage and pricing power that only grows as the company gains even more scale.
The company has consistently beaten its guidance. In early 2011, EQIX said that 2013 revenue would be greater than $2b when the Street was 9% below this figure. Consensus for 2013 is now at $2.25b and (as I will argue below) is still too low. EQIX is now guiding for 2015 revenue of at least $3billion. Like in early 2011, the Street remains skeptical and is well below management's long term guidance. Equinix is planning to convert to a REIT in 2015. REITs are typically valued using Adjusted Funds From Operations (AFFO), which is a proxy for operating cash flow less maintenance capital expenditures. Other datacenter REITs like Digital Realty (DLR) currently trade at about 16x 2013 AFFO. Wireless tower companies like American Tower (AMT) trade at over 20x AFFO. I estimate that at $3b of sales, Equinix can generate AFFO of at least $21 per share in 2015. At 15x, this would imply an end of 2014 target of at least $315 or over 64% upside from current levels.
The company guided for Q3 revenue of $492-498m and the Street is currently modeling $494.5m. This represents a sequential increase of $16.5m versus Q2 excluding the impact of acquisitions. This compares to the $14m sequential increase in revenue from Q2 versus Q1. The Street does not appear to be modeling much improvement in new revenue generation in Q3 versus Q2 despite several positive changes:
- Lower churn. Last Q, customer churn was higher than normal at 3.2%. Most of the increase in churn was because of Equinix's "optimization efforts." In hot markets, Equinix tries to proactively churn out older customers who may be less profitable. It then quickly sells out the newly available space to higher paying customers. Eqinix's CFO stated on its last quarterly conference call that churn would "moderate down to between 2.4% and 2.8%" in the 2nd half of the year. I estimate that a 0.6% lower churn rate implies at least $2m of incremental revenue in Q3 versus Q2. Moreover, because some of the vacated space will likely be quickly filled as per the company's plan, Q3 should have an additional $1m+ or so revenue from new bookings associated with that available inventory. The company has done these optimization efforts in the past, and each time the vacated space was sold very quickly to much more profitable customers. On the Q2 conference call, President of North America Charles Meyers stated, "Typically, we are able to resell capacity quite quickly… these are into high-demand assets with high levels of fill in."
- Significant capacity additions and sales force productivity. Equinix expanded its sales headcount by about 50% over the last 18 months. The company said that bookings last Q were the 2nd highest in its history. Q1 bookings were the highest. Nevertheless, the sequential growth in recurring revenue was actually higher in past quarters. I believe the reason for this discrepancy is because some of the bookings in Q1 and Q2 were for pre-sales of space in new datacenters. On the EQIX's investor relations page, the company offers a PDF that details its expansion plans. 7 new datacenters were scheduled to open in Q3 making the quarter one of the most important for capacity expansion in the company's history. I don't remember a time when this many new data centers opened in a single quarter. To put this into even more context, the same sheet only shows that only 2 new data centers opened in the entire first half of 2012. It takes time for new sales people to become productive and it appears that Equinix may have perfectly timed productivity increases from new sales people hired over the last 18 months with the significant new capacity. Even modest assumptions of new sales associated with this new capacity should lead to upside.
- Currency. When the company gave guidance at the end of July, management said they expected currency to be a sequential drag of about $4m in Q3 after being a $3m sequential headwind in Q2. At the time, the Euro was 1.21 to the Dollar. It currently is over 1.30. Other currencies have also appreciated. I estimate that instead of being a $4m sequential headwind, currency will actually be a slight tailwind in Q3. Therefore, all other things being equal, currency effects alone would imply over $4m revenue upside.
- Acquisitions. In July, the company closed a couple of important strategic acquisitions in markets outside of North America. Equinix's CEO stated that Q3 would have about $13-15m of revenue associated from these acquisitions. However, this figure does not take into account cross-sell opportunities. Many of Equinix's customers do business globally. When the company made foreign acquisitions in the past, it was able to sell some North American capacity to newly acquired foreign customers. It was also able to sell the newly acquired foreign capacity to existing Equinix customers. "Revenue synergies" take some time to materialize, but it is another positive variable for Q3 that was not in Q2.
Given the factors above, I am fairly confident in Equinix delivering short term revenue upside. And, given the very high operating leverage of this business, even small top line upside can yield very large upside in earnings. Moreover, due to the high recurring nature of Equinix's revenue, any upside in the short term would flow through to upside for future quarters. Finally, improved sales force productivity and a higher pace of capacity additions will not be unique to Q3. Q4 is expected to have 5 new data centers and Q1 is also expected to have 4 new centers. This new capacity combined with synergies from recent acquisitions and lower churn will likely drive an acceleration of growth over the next few quarters.
A stock price is a function of earnings and the multiple applied to its earnings. Earnings estimates are likely to be revised upwards and, in my experience, growth stocks that show accelerating fundamentals typically get rewarded with improved earnings multiples.
Risks to monitor are changes in the company's schedule for new datacenter openings, currency movements and competitive industry pricing. There is also the risk that the IRS will deny Equinix's application to become a REIT although I think the probability is low and the IRS is unlikely to even comment until the middle of 2013.
Disclaimer: The author may make trades in securities mentioned without notification. The information contained in this article is impersonal and not tailored to the investment needs of any specific person. You should consult with a professional where appropriate. The author shall not be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.
Disclosure: I am long EQIX.