Mstar of Equinox

Equinix's Adherence to Its Colocation and Interconnection Model Distinguishes It From Competitors

Matthew Dolgin

Equity Analyst

Business Strategy and Outlook | by Matthew DolginUpdated Feb 28, 2019

Equinix is the largest provider of colocated data centers in the world, and it has developed network-dense locations in major cities that would be extremely difficult for competitors to replicate. Telecom networks, cloud service providers, and other enterprises house their equipment and connect with each other at Equinix locations, and the presence of each attracts the others. We expect the importance of interconnection will continue to grow, and in our view, no company is better positioned to take advantage than Equinix.

The utility of colocated data centers, where many different enterprises house their servers and networking equipment under one roof or on one campus, is growing as companies rely more on hybrid cloud models, where they use a combination of cloud services and their own hardware. By colocating, enterprises can directly connect their own equipment with multiple cloud providers, resulting in reduced latency and superior security, making the data center a draw for both sides. Equinix counts all the major cloud providers as clients, and, with other IT service providers, they generate about 30% of Equinix's revenue.

While we think the cloud will be the biggest source of future growth, we think network service providers are crucial in enabling that growth. Not only are networks necessary for Equinix's tenants to reach the outside world, but they use Equinix facilities to peer with each other, making the locations critical components of the Internet's backbone and providing almost a quarter of Equinix's revenue. With its presence in many of the most traffic-dense cities in the world and more than 1,700 network service provider customers, Equinix facilities are major Internet exchange points in many of the world's biggest cities. We think it would be virtually impossible for those locations to move.

While Equinix's model puts the company in the sweet spot today, the asset-heavy nature makes it susceptible to a downturn in data center demand. If the necessity of cloud service providers to the model drives those companies to become competitors or technological advances lead to tenants needing less space, Equinix could find itself with too much capacity.

Economic Moat | by Matthew Dolgin Updated Feb 28, 2019

We assign Equinix a narrow economic moat based on a network effect that results from its customers' needs to connect with each other and switching costs. These advantages make it unlikely for its customers to consider price as their foremost concern in choosing to initiate or continue a relationship with Equinix. Equinix has averaged a 12% adjusted return on invested capital, or ROIC, annually over the last 10 years, and we expect the annual average to be similar over the next decade.

In our view, the most important advantage Equinix has is the network effect, which directly results from its catering to a larger number of companies within each of its data centers. Equinix tenants can connect with each other directly within a site or across Equinix sites globally without having to use the public Internet, which offers tenants superior security and, in many cases, reduced latency, which can be vital. Equinix has nearly 10,000 customers, including 46% of the Fortune 500 and 33% of the Forbes Global 2,000, and we think the presence of these companies makes Equinix data centers attractive landing spots for other companies. While we think the network effect has been a crucial part of Equinix's success and drives the rental revenue it receives for leasing space, Equinix is also able to directly monetize it. In 2018, Equinix generated 16% of its revenue from interconnections, which result from the company charging a fee to directly connect enterprises to each other within data centers or across Equinix data centers.

Perhaps the most crucial of Equinix's clients in driving the network effect are the network providers, which are necessary to carry data from the data centers to the outside world. Enterprises are dependent on networks and are therefore drawn to data centers where many networks are connected, and network providers are most attracted to data centers that have many enterprises, which provide a large potential customer base. The result is a cycle that makes the data center increasingly attractive. Additionally, the presence of so many networks (Equinix has more than 1,700 connected to its data centers) makes data centers excellent Internet exchange points, where networks connect and exchange traffic with each other to form the backbone of the Internet. Network dense data centers--data centers with many networks connected to them--are extremely valuable.

We also think Equinix benefits from the significant switching costs present for tenants that might otherwise prefer to move from one data center to another. We think that once tenants choose an Equinix data center, they are locking themselves into that data center for the long term in most cases. While tenants obviously leave if they no longer have a need, such as if a firm has been acquired, we think it is very uncommon for a tenant to move from one provider to another for the same geographic need. Further, we think it is unlikely that price is ever the motivator if they do choose to move, because we think the costs to move and the risks they take would negate any potential monthly rent savings. According to InfoTech Research Group, the cost to move one cabinet's worth of equipment is $10,000. For context, Equinix's monthly recurring revenue per cabinet is about $2,000, and retail contracts typically last two to four years. Given the lack of long-term certainty in moving to a new data center (due to the relatively short contract), we think firms are unlikely to transfer data centers absent serious quality concerns. For example, assuming a 10% savings versus Equinix's average rate, a customer would face a 50-month payback, which is beyond the length of a typical contract. Beyond the monetary cost, moving servers that are critical to businesses' operations come with risks. Down time alone can be harmful and expensive, and firms often have to engage in mirroring to mitigate it, which is itself costly. We think it is unlikely firms move to competitors once they are in data centers, and in instances they do, we think it is unlikely based on price, which we think gives Equinix given its already strong customer base.

We are convinced that the current environment results in a sustainable competitive advantage that makes it likely Equinix will earn excess returns over the next decade, but the evolving nature of technology keeps us from labeling Equinix a wide-moat firm. On balance, the advent of the cloud has been good for data center businesses, as the number of firms reducing their own data center needs has been much more than offset by those that increasingly move off-site for interconnection with cloud service providers for their hybrid cloud needs. However, the speed with which technology changes and data centers' close connection to it leave us wary about making a twenty-year assumption with any confidence.

While we see the possibility that enterprises will have less need for data centers if they use infrastructure as a service, or IaaS, to a greater extent, our bigger concern is that companies that don't make that move will be able to satisfy their needs with less space. Hardware with more capacity in a smaller physical size and virtualization could each reduce the need for space over the long-term even as enterprises use data centers for an increasing amount of data. Also, as cloud service providers become even more dominant, we expect they will have increasing control over data centers' fortunes. While they seem to be great customers now, we will not bank on that considering the leverage that they hold.

Fair Value and Profit Drivers | by Matthew DolginUpdated Aug 02, 2019

We are raising our fair value estimate for Equinix to $420 from $415. Our valuation multiple implies an EV/adjusted EBITDA multiple of 17 and a price to AFFO multiple of 19 based on our 2019 estimates.

We project top-line revenue growth to average about 8% annually over the next decade. Our forecast assumes cabinet additions gradually decline as the market gets more saturated, but spending needs will come down in turn. We project the firm will add about 30,000 cabinets in 2019 and that new additions will gradually decline to around 12,000 by 2026 before leveling off. We believe less incremental capacity will cause utilization to creep up—from around 80% in 2018 to 83% by 2028. We expect revenue per cabinet to rise only modestly—about 2% annually—as we think Equinix's reliance on its major customers will keep a lid on its pricing power and offset higher escalators for other enterprise customers. We think interconnection will continue to be the fastest growing portion of Equinix's business, driven mostly by customers having continually greater needs to connect with their cloud service providers. We forecast interconnection revenue to average close to 9% growth annually through 2028.

We expect spending needs to remain high in the near term as Equinix continues expanding cabinet capacity and moves to own more of its data centers. We expect capital spending in 2019 to be about $2 billion for the second straight year. We project capital expenditures to gradually decline beginning in 2020 as cabinet expansion begins to moderate. We believe Equinix can expand its adjusted EBITDA margin as it gradually shifts to owning more of its property, reducing rent expense. We forecast adjusted EBITDA margin to rise more than 500 basis points over the next decade from 47.6% in 2018.

Risk and Uncertainty | by Matthew Dolgin Updated Feb 28, 2019

Equinix's uncertainty rating is high, primarily because the capital-heavy nature of its business minimizes its ability to be agile in the face of a market downturn. If demand for third-party data centers weakens for any reason, Equinix would find itself with too much capacity and a weakened cost profile.

We think technological advances pose the biggest risk to Equinix's long-term prospects, as they could result in an environment where cloud growth, interconnection, and data usage continue to grow, but Equinix is not the beneficiary. Virtualization and Moore's law could result in a scenario where Equinix's customers satisfy their exploding data needs with less physical space. Similarly, increases in the amount of data companies can push through a given fiber connection result in fewer necessary cross connects for the same amount of traffic. A long-term faster pace for advances in capacity relative to data connection needs could weigh on interconnection revenue. Finally, a greater shift by enterprises toward IaaS cloud adoption could reduce their own need for server space, as their data is shifted to cloud providers' servers and stored more efficiently. While we think the latter example is less likely en masse because we think large enterprises value the security and privacy of keeping proprietary hardware, it could happen at the margins.

Power becoming concentrated in the big cloud providers is the other big risk we see for Equinix. Cloud and IT service companies account for 30% of Equinix's revenue, but more important, the group is a major draw in attracting other tenants. We expect cloud providers' importance to continue growing, leaving Equinix more susceptible to these companies playing hardball. We don't think cloud providers want to compete with Equinix's colocation business, considering the difficulty we believe they'd have in mimicking a similar network of their customers, but we think their leverage could result in less favorable terms for Equinix.

Stewardship | by Matthew Dolgin Updated Aug 30, 2019

We assign Equinix an Exemplary stewardship rating. Over its two-decade history, the company has consistently shown the foresight to take advantage of burgeoning trends. The company was founded in 1998, before the Internet was as ubiquitous as it is today, with the vision of providing carrier-neutral Internet exchange points. It has since amassed over 1,700 networks throughout its 200 worldwide data centers, and it owns many of the most highly trafficked Internet exchange points in the world. Though it was founded with the simple purpose of connecting networks, it evolved to a colocated data center company that houses servers for thousands of companies and connects them with each other. We think this move was crucial in positioning itself to profit from enterprises' reliance on the cloud. Equinix has historically been one step ahead of the trends, and we credit visionary leadership.

Charles Meyers has been CEO since September 2018. He joined Equinix in 2010; became the company's chief operating officer in 2013; and was most recently Equinix's president of Strategy, Services, and Innovation, before taking the top job. Prior to joining Equinix, he held leadership roles with multiple companies in the information technology industry in a career that has spanned more than 25 years.

Equinix's chairman since 2007 has been Peter Van Camp, an industry veteran who served as Equinix's CEO from 2000-07 and again on an interim basis for most of 2018. Van Camp has been with the firm for almost its entire history.

We believe Equinix's corporate governance is friendly for shareholders. Except for Van Camp's stint as interim CEO in 2018, the chairman and CEO roles have always been split, which we think is important for good oversight and less concentration of power. Additionally, all directors are elected on an annual basis, allowing sufficient control for shareholders, and seven of the nine current directors are independent. Compensation for the company's top five executives consists mostly of long-term incentives (about 80% of total compensation), with the rest split equally between base salary and annual incentive compensation. We think the compensation amount and structure is very fair for a company of Equinix's size.