CyrusOne Inc.
Q2 2018 Earnings Call Transcript

Published:

  • Operator:
    Good day and welcome to the CyrusOne Second Quarter 2018 Earnings Conference Call. All participants will be in listen-only mode. After today's presentation, there will be an opportunity to ask questions. Please note this event is being recorded. I would now like to turn the conference over to Michael Schafer. Sir, please go ahead.
  • Michael Schafer:
    Thank you, Steven. Good morning, everyone, and welcome to CyrusOne second quarter 2018 earnings call. Today, I am joined by Gary Wojtaszek, President and CEO; and Diane Morefield, CFO. Before we begin, I would like to remind you that our second quarter earnings release along with the second quarter financial tables are available on the Investor Relations section of our website at cyrusone.com. I would also like to remind you that comments made on today's call and some of the responses to your questions deal with forward-looking statements related to CyrusOne, and are subject to risks and uncertainties. Factors that may cause our actual results to differ from expectations are detailed in the company's filings with the SEC, which you may access on the SEC's website or on cyrusone.com. We undertake no obligation to revise these statements following the date of this conference call, except as required by law. In addition, some of the company's remarks this morning contain non-GAAP financial measures. You can find reconciliations of those measures to the most comparable GAAP measures in the earnings release, which is posted on the Investors section of the company's website. I would now like to turn the call over to our President and CEO, Gary Wojtaszek.
  • Gary J. Wojtaszek:
    Thanks, Schafer. Howdy, everyone, and welcome to CyrusOne second quarter earnings call. This was a tremendous quarter, especially for the sales organization as the company set numerous leasing records. Demand is as strong as it's ever been, and we are in a very good position as we head into the second half of the year. As slide 4 shows, we put up great results across the board. Revenue of $196.9 million was up 18% over the second quarter of 2017, and adjusted EBITDA of $110.6 million was up 22%. Normalized FFO was up 19% and FFO per share was up 5%. The box in the upper right section of the slide highlights the leasing records for the quarter with 52 megawatts and 305,000 colocation square feet signed, totaling $65 million in annualized GAAP revenue. We also set several other leasing records, which I will discuss on the next slide, and backing our $85 million of annualized revenue is the highest quarter in total in company history. To support the leasing, we have a substantial pipeline consisting of more than 400,000 colocation square feet and 86 megawatts of power capacity under development. We also acquired 68 acres of land in Mesa, Arizona in the second quarter, and we announced last week we acquired a shell in Northern Virginia with up to 33 megawatts of power capacity. These are two of the strongest data center markets in the country and among our best performing markets. Internationally, we are continuing to focus on establishing in European footprint. In addition to the presence we will have in London and Frankfurt once the Zenium acquisition closes, we have additional sites in process in those two cities as well as in Dublin and Amsterdam that would increase our capacity in these markets to nearly 250 megawatts. We also completed construction of a 350-foot communication tower at our Aurora I location during the quarter and we will be offering the first on-campus wireless access to customers. As slide 5 highlights, this was a phenomenal leasing quarter, and as I just mentioned, we set a number of records. The revenue signed was more than double the prior four-quarter average, and through the first half of the year, we have signed more revenue than we did all of last year. Our revenue signed was more heavily weighted toward the cloud vertical than in recent quarters, including two Chinese cloud customers. And as I mentioned on last quarter's call, the development yields in these larger multi-megawatt deals are in the low teens range. We also signed more than 500 leases for the first time in a quarter in our history. And consistent with what we have said in recent quarters, it was broad based, covering nine verticals across nine markets. The number of leases signed was nearly 20% above our prior four-quarter average. The weighted average term of the leases signed was also a quarterly record at nearly 12 years. And the weighted average remaining lease term of the portfolio is nearly five years, the longest in the company's history. Customers have been increasingly requesting longer-term leases to give them more certainty and the extended portfolio duration enhances the durability of the cash flow stream. Lastly, we signed $2.8 million in interconnection revenue, also a quarterly record for the company. Total interconnection revenue was up nearly 20% in the quarter and we had another 1,300 cross-connects. The interconnection growth continues to be driven by the development and growth of ecosystems within our data centers as well as demand for SDN-enabled cloud interconnection. In short, it was an incredible quarter. We signed a record number of deals, the highest monthly revenue with the longest weighted average lease duration in our company's history, and had the strongest interconnection bookings ever. Great job, Tesh. She clearly earned this keep this quarter. Slide 6 highlights the joint efforts of the sales and construction teams in driving our growth. As you know, we view our sales force as one of the key value drivers of the business, and the investments we have made over the years to build a world-class organization have really paid off. Among the data center REITs that report quarterly revenue signed, through the first half of 2018, we have accounted for 34% of total bookings, which is double our revenue share of 17%. So, we are punching well above our way. The revenue signed over the last four quarters represents 23% of our base business, implying very strong organic growth. As you know, we have expertise in selling to the Fortune 1000, have been very successful penetrating hyperscale segment, with the business model now representing roughly one-third of our portfolio. Additionally, the credit quality of our customer base is better than you would typically find in most REIT asset classes and over 70% of our rent is from customers with investment-grade credit ratings. All the leasing success we have had would not be possible without the efforts of our construction team. Our ability to bring significant amounts of capacity online quickly and cost effectively is also a key value driver. Upon completion of the projects in our development pipeline, we will have delivered approximately 130 megawatts in just over a year, increasing the size of the footprint by nearly 25%. We also have significant runway for future growth. Between what's currently available and under development, we have approximately 500,000 colocation square feet available for lease, which we estimate will generate up to $125 million in incremental revenue upon full lease-up. Over the past few quarters, our late-stage sales funnels have continued to increase. With the record leasing quarter this quarter, it's down 4% sequentially. However, it is up 23% compared to a year ago and near its highest historical level. So, the inventory we have positions us well to meet this demand. Additionally, with the shell and land we have in our portfolio, we can triple the size of the company any more than a gigawatt of power capacity. Turning to slide 7, I am very excited about the tower we built at Aurora I facility just outside of Chicago. As most of you know, this is the data center that we acquired from the CME in early 2016. It houses the premier electronic trading platform, providing global connectivity to futures and options across all asset classes. The tower creates the first on-campus wireless access and offers an equal access for all of our financial ecosystem customers. We are currently in discussions with potential customers and expect to begin generating revenue by the end of the year. We expect to achieve very high returns on what is a nominal investment made possible because of our ownership of the data center. We are constantly trying to think of creative ways to further monetize our assets and this is just another example of what we do. Slide 8 provides an update on our expansion plans in Europe. Demand continues to grow across the continent and there was an estimated 36 megawatts of absorption in the first quarter, up 35% versus a year ago and 20% above quarterly average for the full year. The trend toward larger deployments is continuing, which is consistent with the discussions we are having with numerous hyperscale and enterprise customers about potential solutions across these locations. Zenium is on track for continued strong growth with a sizable revenue backlog and a robust sales funnel. As was the case when we reported first quarter earnings, we are just awaiting the German regulatory approval to close the acquisition. While that process is obviously taking longer than we had hoped, we anticipate that we will receive approval in the next couple of months, and this will give us a presence in London and Frankfurt, the two largest data center markets in Europe. We also have sites in process across London, Dublin, Frankfurt, and Amsterdam with additional potential capacity of approximately 200 megawatts for a total prospective footprint of nearly 250 megawatts. By 2020, we will have one of the largest platforms in Europe. Moving to slide 9, as a direct result of our strategic partnership with GDS, during the second quarter, we signed leases totaling more than 10 megawatts, with two Chinese hyperscale customers. In addition, we are in discussions on approximately 25 deals for deployments in the U.S. or China as a direct result of the partnership. Additionally, we're being contacted by Chinese hyperscale companies, looking to expand in Europe, which further validates our decision to expand internationally and creates a great incremental opportunity for us as we establish our European footprint. When we initially made the $100 million investment in GDS, we needed to sell about 7 megawatts of data center capacity to make it accretive. As a result of our strategic partnership, we closed 10 megawatts of deals this quarter, making our initial investment accretive on a pro forma basis. As we have seen in our business over the last decade, we know that customers will grow with us over time and we will continue to add more Chinese customers, enhancing long-term value creation. Regarding the report from the short sale on GDS that was published on Tuesday morning, CyrusOne cannot comment, but I encourage any investor who has any concerns about the allegations to reach out directly to GDS to provide further color on the allegations raised and the rebuttal they released yesterday morning. That said, we believe that the initial thesis underlying our strategic partnership with GDS remains intact. We continue to believe that the U.S. and China will be the two countries that will dominate cloud and artificial intelligence, and we needed to position our company for the future as this industry becomes more global. We recognize that China would be an incredibly difficult market to enter into on our own and we wanted to partner with a world class data center company and work with an individual who we trusted and shared a similar vision for our industry, and who had strong relationships with Chinese cloud companies that we can help grow outside of China. Similarly, we wanted to partner with a Chinese data center operator who we felt confident would provide the same level of support and dedication we provide our customers as we help our customers grow in China. Nine months into our strategic partnership with GDS, I can unequivocally say that William and his team have been fantastic partners. It has been my experience that most JVs do not end up working as well as expected, and the root cause is generally a lack of follow-through and commitment by the respective teams to do what they said they would. In our case, GDS has done exactly what they said they would. The CyrusOne teams have visited China several times and the GDS teams have come to the U.S. several times. The underlying relationships that have developed between the teams is very strong and evident to everyone involved in this process, which is a testament to the commitment both companies have toward our common objectives. One of my measures of a great partner was doing business with a person who does what they said they would do, and William has done exactly that. He is a great entrepreneur, who has built an incredible business, and I'm sure that the relationship we have built with the GDS team will further improve over time. In closing, we are off to the strongest start of the year since going public. We had record bookings this quarter and sold more in the first half of this year than we did all of last year. This is all against the backdrop of demand – the biggest backdrop in demand that we have ever seen with record industry bookings for the quarter, which were up 50% compared to last year. I do not recall a period of time when the industry and our performance has ever been stronger. We have a substantial amount of inventory available to meet this demand and our international expansion efforts should allow us to continue to grow well beyond 2020. And we'll now turn the call over to Diane, who'll provide more color on our financial performance for the quarter and an update on our guidance for the year. Thanks.
  • Diane M. Morefield:
    Thank you, Gary. Good morning, everyone. As Gary noted, we posted very strong results for the quarter. Turning to slide 11, our revenue, adjusted EBITDA, and normalized FFO, all grew in the high teens to low 20% range. The impact of the Sentinel acquisition is now fully reflected in our year-over-year comparison. So, our growth this quarter was entirely organic. Consistent with recent quarters, churn remained low at 1.1%. However, as I mentioned on last quarter's call, we expect churn for the full year to still be in the 6% to 8% range based on some significant known churn related mainly to non-renewing leases that will occur in the second half of the year. We'd like to highlight when comparing total revenue between the second quarter this year over the first quarter of this year, the results were relatively flat. However, the reason for this is the $5 million lease termination fee we recorded in Q1 as compared to zero lease term fees this quarter. Turning to slide 12, NOI grew 19% in the second quarter, driven primarily by the increase in revenue over last year. The adjusted EBITDA margin was up 1.8 percentage points to 56.2%, again driven largely by revenue growth without a commensurate growth in overhead. Normalized FFO grew roughly in line with adjusted EBITDA and normalized FFO per share increased 5%. As the chart at the bottom of the slide shows the net impact of the adjustment to normalized FFO to arrive at AFFO was fairly muted this quarter at approximately $2 million, which is less than the run rate in the prior two quarters as a result of lower straight line rent adjustments in the second quarter. Slide 13 provides an update on the portfolio by market. We have a broad domestic footprint with data centers in major markets across the country, providing an attractive, comprehensive solution for our customers. And 70% of our revenue is from customers in multiple locations throughout our portfolio. Our expansion into Europe will further enhance the value of our platform and diversify the portfolio. Total colocation square feet, which I will refer to as CSF, leased and CSF leased within our stabilized properties, were basically in line with last year's percentages. Additionally, total CSF leased was up 2 percentage points from the end of the first quarter, and 5 percentage points from the end of last year, representing the conversion of our late-stage sales funnel into bookings during the first half of the year. As Gary mentioned, our construction team continues to do an outstanding job, delivering capacity to support our sales team, bringing approximately 800,000 CSF online over the past year. During the quarter, we exited two very small properties, our Marsh facility in Dallas, and our Goldcoast facility in Cincinnati. The Marsh facility was a leased property that we chose not to renew, and our Goldcoast facility was an older owned property, in which the last customer had moved out. The related rent impacts are captured in our second quarter churn. Slide 14 summarizes our development pipeline as of the end of the quarter, which includes projects in Dallas, Northern Virginia, and the New York Metro area that will deliver approximately 400,000 CSF and 86 megawatts of power capacity. As a result of the strong leasing, 78% of this square footage under development is preleased. This is the second highest preleasing percentage in the company's history and is reflective of dynamics more broadly across the industry with customers willing to sign leases before the capital has been deployed to build the data halls. While we have always built our data halls on a just-in-time basis to minimize our capital at risk, this high level of preleasing further de-risks our pipeline. Upon completion of the projects in the pipeline and the closing of the Zenium transaction, the size of our portfolio will be more than 4 million colocation square feet. Slide 15 highlights the strength of our balance sheet and credit quality of the company. Our leverage remains low at 4.7 times as of June 30. We have no debt maturing for five years, a weighted average remaining debt term of more than six years, and a fully unencumbered asset base with a gross book value over $5.5 billion. On a market cap basis, our debt represents only 28% of our capital structure, with equity representing 72%. And at the end of the quarter, we still had more than $2 billion of liquidity to fund our growth opportunities. During the quarter, Moody's upgraded our corporate credit rating to Ba2 and maintained their positive outlook. While we were pleased with the upgrade, we continue to believe we should be an investment-grade company as we have highlighted in the past with our benchmarking to other REITs that have investment-grade rating. We will continue to maintain an active dialog with the rating agencies to help them gain a better understanding of the data center industry and our conservative financial policies. Our fixed to floating rate mix is in the 50%/50% range, and upon achieving an investment-grade rating, we would expect to weight it more heavily towards fixed rate debt. Turning to slide 16, our revenue backlog as of the end of the quarter was $85 million. As Gary mentioned, this is the highest quarter in backlog in the history of the company. Given our estimated timing of commencements, this backlog will have less of an impact on this year's results, but does provide a strong baseline for continued growth into 2019. As shown in the bottom of the chart, we anticipate nearly two-thirds of this revenue will commence over the next two quarters. Moving to slide 17, we are updating our guidance for 2018. With the exception of the increase in our metered power reimbursements range, the changes to our ranges are attributable to a change in our assumptions regarding the timing of the Zenium closing. As Gary mentioned, we are still awaiting German regulatory approval and we are now assuming an October 1 closing date. Our previous guidance had assumed that we would close in early May. As a result, we are decreasing the ranges for leased and other revenues from customers and for adjusted EBITDA to reflect this five-month lag in closing Zenium, which is not in our control. We are increasing the range for normalized FFO per share to reflect the positive impact associated with the delayed timing of funding the acquisition. The increase in our metered power reimbursements range is driven primarily by a faster ramp in metered power that we had originally – faster than we had originally anticipated. As you know, this has no impact to adjusted EBITDA as these reimbursements are associated with pass-through power contracts. We are reaffirming our guidance for capital expenditures in the $850 million to $900 million range for the year. In closing, it was an outstanding quarter with a number of notable accomplishments for CyrusOne. Given our tremendous leasing success in the first half of 2018 and our strong sales funnel, we feel really good about our position to continue to achieve strong growth in 2019 and beyond. In addition, the secular demand drivers continue to provide an incredible tailwind to the broader data center industry in the coming years. We thank you for participating in the call and we are now happy to take questions. Operator, please open the line.
  • Operator:
    Thank you. We will now begin the question-and-answer session. And our first question comes from Frank Louthan with Raymond James. Please go ahead.
  • Frank Garreth Louthan:
    Great. Thank you. Wanted to get some comments little bit on the average contract length and development yields for the deals that are being booked. Walk us through why those are extending a little bit? And then as it relates to the other 25 Chinese deals you talked about in the slide, are any of those here in Europe and is there any concern that some of the U.S. trade relations might impact these or stretch them out of it? Thanks.
  • Gary J. Wojtaszek:
    Sure. Hey, Frank. Yeah, so, it's kind of like what we've been saying for a while. I mean the length of the lease term is really kind of a good indication of just how much more comfortable all of our customers are with outsourcing their data center needs to us as a replacement to doing it on their own. We expect that that's going to continue to go out. I mean our longest lease duration in there was about 15 months. So, it's skewing longer over the years and you kind of see that in the average lease duration of our portfolio, which is about 60 months now, so really long-term. In terms of the yields, as I mentioned on last quarter's call, for some of the bigger cloud deals, we're looking at low teens, so around 13% type yields, so about 200 basis points below the 15% that we have typically been generating over the year. So, if you look at it on a blended basis, the smaller deals are still averaging at the higher rates. But proportionally, given we did a lot of these big cloud deals this quarter, it would skew more towards to the lower end of that range, but it was a pretty strong quarter overall.
  • Diane M. Morefield:
    And from a REIT land, if you look at low to mid-teens unlevered returns, those equate to 30% range levered. So, in terms of any other REIT product type and even in terms of our returns to some of our competition, those are really off the chart even with a bit of compression in the unlevered returns for the cloud deals.
  • Gary J. Wojtaszek:
    Yeah. And with regard to the 25 deals that are in the funnel, those are – in all of my commentary, when we talk about funnel and what that look like, that's exclusive of Europe. And so, we actually have a presence there that we can really sell into. We don't want to include that in there. So, in aggregate, our sales funnel is down 4% sequentially up over 20% versus last year. Still, really, really strong. The 25 customers that are in the funnel for China, those are either in for our customers looking to go to China or additional Chinese customers looking at coming to the States. I don't think it's going to really impact the whole trade war issue that's going on because this is really – we're kind of just supporting data flows. And that to date has not really been that impacted by that. I don't know if you saw earlier this week, but Google was announcing that they are going to be launching a cloud offering – I mean a search offering in China as well.
  • Frank Garreth Louthan:
    Great. Thank you very much.
  • Operator:
    Our next question comes from Robert Gutman with Guggenheim Partners. Please go ahead. Robert, your line is open.
  • Robert Gutman:
    Hi. Thanks for taking the question, and great leasing quarter. I was wondering if you could talk a little bit more about the impact of the tower in terms of a little more about the economics and the revenue-generating potential, and what does it mean in terms of further development of another facility in the area?
  • Gary J. Wojtaszek:
    Sure. Hey, Rob. Yeah, so, the tower is something that we've been working on for over a year. So, our facility in Chicago is and has been the epicenter for all the futures trading. I think there's something like $25 quadrillion of notional trading volume that goes through that facility per day, driven by the CME's exchange being located there. So, it's one of the primary financial hubs in the country and speed is of the essence there. That's really kind of ground zero when you're talking about kind of high speed trading, basically trying to arc the difference between the New York spot and futures market in Chicago. So, you're trying to get those bits and bytes to move across the country as fast as possible. This tower, which will be the largest tower in that area, will enable those customers on that to access that high speed to basically be as fast as possible into the matching engine in Chicago. We're going to be pretty broad and open to that to offer it to as many customers as possible. The economics on this are very compelling just given how unique an asset it is. So, we expect that, in aggregate, it's not going to be that impactful to our aggregate numbers overall. I mean maybe you get a couple of million bucks of revenue associated with this thing. But the more important aspect of this is really just to highlight that this is just another example of how we look at our portfolio and try to monetize as many assets as possible. I mean we were the market leaders and always focusing on distributor – redundant system enables us to get really high yields in our data centers. We focused on low resiliency solutions that enabled us to get even higher yield on some of the assets that we're deploying. We've done a number of these things over the years to basically try to drive as much return out of the capital that we're putting in the ground and this is just another example of that.
  • Robert Gutman:
    Great. Thank you.
  • Operator:
    Our next question comes from Simon Flannery with Morgan Stanley. Please go ahead.
  • Simon Flannery:
    Great. Thank you. Good morning. Nice quarter on the leasing. Perhaps, Gary, you could just expand a little bit more. We've seen obviously strong results from you, strong results from most of your peers, and really talking about this big pickup in activity first half of 2018 over first half of 2017. So, you called out some of the Chinese cloud demand. Can you just give us a little bit more color into the leasing and the pipeline and the verticals, the regions? What is it that we're seeing here and how sustainable is it for you and for the broader industry?
  • Gary J. Wojtaszek:
    Yeah, look, I mean the commentary on the broader industry is really good because there was so much furor I think coming into the first quarter, everyone concerned about what's happening, and I think what you saw last quarter, everyone put up big numbers. Some of my peers put up record quarters for themselves last quarter. I think you see the continuation of that this quarter with everyone doing really, really well. So, I think that's just a validation of what we've been saying for a while. I mean after the last earnings call, I talked about meeting with three of my customers, each talking about needing a gigawatt of capacity each. So, with three customers talking about needing 3 gigawatts of capacity over the next five years, it gives you a really good indication of what's happening in this industry. So, we feel really confident that the demand trends in this industry are really broad, really big, and will continue to accelerate over the next several years. With regard to our funnel or the sales this quarter, maybe start there, it was broad based. We sold to nine different markets across nine different industry verticals. We had four different cloud companies included in there, and a number of additional enterprise companies. So, it was very strong, and we expect that this is going to continue over time. With regard to our aggregate funnel, it was down 4% sequentially, up 23% year-over-year, still pretty close to our near record big funnel for us. And so, we expect that as we look ahead into the second half of this year, into 2019, we feel really good about where we're positioned, particularly with a backlog that's over $80 million right now. So, we expect this is going to continue, and we expect that when we get into Europe, we're going to further be able to accelerate our velocity here because right now, we're basically just a U.S. player, and we feel that we have a lot more potential to sell broadly internationally, and we just need to build out that platform to be able to attack those other markets.
  • Simon Flannery:
    And any comment on what's holding up the Zenium deal?
  • Gary J. Wojtaszek:
    No. We've been in this holding pattern with the German regulators for a while. I mean, they passed a new law and made the whole GDPR thing. There's a lot of disconnect between how the law was passed and how the various institutions in Germany are going to implement that law. So, we're unfortunately in that position right now that we have to work through this. So, we expect that, we push this out to October 1. We continue to work with the German regulators. Just given what happens in Europe in August, we do not expect that's going to happen very, very quickly.
  • Simon Flannery:
    Great. Thank you.
  • Operator:
    Our next question comes from Sami Badri with Credit Suisse. Please go ahead. Sami Badri - Credit Suisse Securities (USA) LLC Thank you. My first question is on the 2Q 2018 lease signings and the average price points they were signed at. Given pricing has come down quite a bit to the $105 per KW level which is about 9% down quarter-on-quarter. And what I really want to know is this the right level we should think about when you work with hyperscalers? And maybe just looking out in the future, is this the right level we should think about for 2Q 2019 or how should we think about price sensitivity?
  • Gary J. Wojtaszek:
    Hey, Sam, thanks. Thanks a lot. It was a really strong quarter. We don't think about price. We think about returns. And because price is just a misleading indicator of what your aggregate returns are. So as I mentioned earlier, we're still targeting a 15% return overall outside of the cloud companies. With the cloud companies, we're targeting around 13% return. We expect that we'll be able to do that for a while. So, the reason why price is not that informative is that there are many things that we do when we're delivering these solutions to our customers that enable us to pull out capital from what we're designing in. So, the price is lower than what you've seen historically but that's also because commensurately the capital that we're deploying is also lower. That said, when I mentioned on the last quarter call and what we're still maintaining is that we do see the yields coming down about 200 basis points from where they were historically. We have seen some creep up in construction costs from the beginning of the year in labor and some of the supplies on that. And so, a lot of the savings that we had engineered on our first solutions from last year's effort, we're kind of giving netbacks, we're kind of holding neutral from a cost per delivery perspective. But in aggregate, what we deliver for hyperscalers is a different solution that we're delivering for the typical enterprise. Sami Badri - Credit Suisse Securities (USA) LLC Got it. Thank you. Very helpful. And then just the second question is more of a follow-up from a prior question that was asked, and this is more to give us more of like the macro view idea. Eight years ago, a 3.5 megawatt lease was considered significant. Today, 35 megawatts is probably considered very significant. But in three years to five years, what are the multipliers from today's level are we thinking? Are we thinking like 250 megawatts to 350 megawatts is what we'll be talking about in four years or five years from today?
  • Gary J. Wojtaszek:
    Well, some of the bigger deals now you're talking about 100 megawatt commitment. So, I would expect a 200 megawatt to 300 megawatt commitment in five years to be probably within the ballpark. Sami Badri - Credit Suisse Securities (USA) LLC Got it. Thank you.
  • Operator:
    Our next question comes from Richard Choe with JPMorgan. Please go ahead.
  • Richard Y. Choe:
    Hi. I wanted to follow up on the leasings. If we separate out the two Chinese deals, 20 megawatt out of the 52 megawatt, what are we looking at in terms of lease terms versus average lease term given that we've gone from 6 years to almost 12 years? How should we think about what the other deals outside of the Chinese deals have been signed?
  • Gary J. Wojtaszek:
    Yes. So, Richard, I'd like to clarify on that. So, the Chinese deal were 10 megawatts. So, it's 10 megawatts out of the 50 megawatts for the Chinese hyperscalers. So, I mean, the average terms range from five years up to 15 years, and that was all across the hyperscalers on the enterprise. Typically, the enterprises are on the lower end of the lease term because there's still – there's two things that they're struggling with. One is they recognize they've got to get outsourced. They've got to get out of their older legacy facilities and take advantage of all the new technologies that are out there. But at the same time, they're very concerned about not wanting to over-commit to go to longer-term because their IT roadmap is really on queer once they get out beyond five years about how much of their IT kit is going to be moving to the cloud versus how much is going to be on a dedicated on-prem basis.
  • Richard Y. Choe:
    And I think you mentioned it before, but just to get a little bit more color on it. I mean, going to an average lease term of 12 years is a long time.
  • Gary J. Wojtaszek:
    Yeah.
  • Richard Y. Choe:
    What's driving that? What – like what are the conversations around why companies are comfortable going to such a long-term?
  • Gary J. Wojtaszek:
    Yeah. So, because the alternative is that they're going to build this facility in themselves, and they were going to own it for 30 years or 40 years. So, they basically want to basically lock in the certainty of having that asset available to them for their use for a really long period of time. And I expect that over time you're going to consistently see leases in that long-term range. I mean some of the deals that we've been talking about with customers, they've been looking for 20-year leases. So, I think that's just an indication of all the customer's willingness to outsource and the fact that they want to basically try to obtain the same type of ownership economics as if they build and manage this facility on their own.
  • Richard Y. Choe:
    And given the strong signs and long lengths of these deals, how should we think about financing the deals?
  • Gary J. Wojtaszek:
    I – well, I mean, in the aggregate, I mean, we – we're going to...
  • Diane M. Morefield:
    Yeah.
  • Gary J. Wojtaszek:
    ...continue to do the same. I mean, Diane can talk...
  • Diane M. Morefield:
    Yeah. I mean...
  • Gary J. Wojtaszek:
    ...more
  • Diane M. Morefield:
    ...as you know, we borrow on an unsecured basis, and we manage our balance sheet holistically. So, again, we have over $2 billion of untapped liquidity where we sit today. Our practice has been once we build, say, up to $400 million to $500 million outstanding on our revolving credit facility, we'll go to the longer-term bond market, and finance in tranches anywhere from 5 to 10 years in the long-term bond market. But today, we have zero out on the credit facility, and we have another $300 million of delayed term loan that we put in place when we closed the new facility at the end of March that we need to draw down this fall. So, right now, we wouldn't go to the bond market because we have nothing to – the bond or equity markets because we have nothing to refinance at this point.
  • Richard Y. Choe:
    Great. Thank you.
  • Operator:
    Our next question comes from James Breen with William Blair. Please go ahead.
  • James D. Breen:
    Thanks for taking the question. Just for Diane on the debt side. You guys got shift to deposit by S&P in May. They talked about leverage levels in the mid to high 5 times. Does it change in any way how you think about financing deals? And then it seems like you're moving in that direction, especially given the growth. Any thoughts on what the potential impact would be to your borrowing costs were that to happen. Thanks.
  • Diane M. Morefield:
    Yeah. Well, I mean we quote net debt to EBITDA on last quarter annualized to our debt outstanding. The rating agencies typically use trailing 12 months. That's why they all show higher leverage than we report. But if you think about it, when you're growing 25% range a year, really the last quarter annualized EBITDA is a more meaningful measure of our actual leverage and that's why we called it that way. So, we're in front of the rating agencies constantly, myself, Gary, Michael Schafer. We meet with them. We have follow-up calls when we announce earnings or when we announce any kind of transaction. Again, we benchmarked and we actually put those slides in our earnings deck last quarter. When you benchmark for REITs our size in the, say, $5 billion to $10 billion market cap range, most of those REITs have over 6 times leverage on net debt to EBITDA. And we have very similar qualities, long-term leases with investment grade credit, rent escalators, et cetera. So, we think we're under-levered quite frankly compared to investment-grade REITs. But we have said we will manage in sort of this low to mid 5 times net debt to EBITDA to continue to have a really conservative balance sheet.
  • James D. Breen:
    Great. And then just any thoughts on what it could mean for borrowing cost if an upgrade were to happen?
  • Diane M. Morefield:
    Oh, absolutely. If you get investment-grade on long-term bonds, it's at least 100 basis points or more delta in the coupon on the fixed rate long-term bonds. Our credit facility margins go down immediately, our borrowing margin under that revolver once we get to investment-grade. I think there's other benefits obviously. The investment-grade market never shuts down in any kind of economic cycle, so you'll always have access to capital. We all know the high-grade market can be much more skittish if there's any kind of sideways economic downturn. So, it's all of those intangibles combined as well as true decrease in your weighted average cost of capital.
  • James D. Breen:
    Great. And then just one for Gary just on the competitive side, obviously a good bookings quarter. Anything you're seeing there in terms of change? Is it still sort of the same competition or are a lot of the clients just coming directly to you and it's not even really a bidding process? It's more like they want you to do the work for them.
  • Gary J. Wojtaszek:
    No. Look, I mean the competition has always been good in the industry. I mean we like competing. We like winning more than losing. And I think we did a good job with that this quarter. But I think what you saw this quarter is everyone did well, right? I mean in aggregate, the bookings for the group was up 50% versus last year and I think that's just giving a broad amount of indication for just how strong this is. We do have just many customers come to us directly where they're not really competitive processes because they know that we will deliver for them and have repeatedly over and over. They like the cost points that we can deliver at and the speed of delivery is something that is incredibly important to them. I talked about previously but in a roughly 20 megawatt facility that we can deliver to a customer that customer will generate about $125 million of revenue. So that's revenue that is available to him if we can deliver. If you miss your deadline, those customers have just lost out some revenue that they will never ever get back.
  • James D. Breen:
    Great. Thank you.
  • Operator:
    Our next question comes from Colby Synesael with Cowen & Company. Please go ahead.
  • Colby Synesael:
    Great. I have some math questions.
  • Diane M. Morefield:
    Shocked.
  • Colby Synesael:
    Yeah, I know. So first off for Zenium, can you remind us or not remind us but tell us what is now assumed from Zenium in the new guidance, so assuming that October 1 close through the end of the year for revenue, EBITDA and core FFO in terms of the dilution? And then secondly, as it relates to financing Zenium, can you remind us how you're intending to finance that? Is that all with debt or is there some form of equity assumed when that deal closes? And just more broadly as it relates to equity, is doing equity between now and the end of the year would you consider that opportunistic or based on your capital requirements in terms of all the builds, it's actually a required component of how you're going to pay for all of these? Thank you.
  • Diane M. Morefield:
    Okay. Let me work through this multi-part question. When we announced Zenium at the end of last year, we showed that bridge in our deck and that just from Zenium it was roughly 13% to 17% dilutive – I'm sorry, $0.13 to $0.17 per share dilutive. So, midpoint would be about $0.15 dilutive. The way we view it, we thought we were going to have about nine months of Zenium, and then the numbers we've lost about five months assuming in our current guidance we're using in October 1 close date. In very round math, it was around $0.02 a month, a little less than that. So, that's where you're losing around $0.10 compared to if we'd closed at April 1. Regarding the funding, again, between the fourth quarter of last year and the first quarter this year, we had raised a significant amount of equity on the ATM in anticipation of closing Zenium. It's why we're at 4.7 net debt to EBITDA, which is the lower 5 to mid-5 range. So, we've really kind of prefunded the equity piece. When we close it, again, we'll draw on the term loan, and the excess on the revolver. So, we will still have significant liquidity on the roughly $1.5 billion range even after we close Zenium. So, that's how we put together the prefunding.
  • Colby Synesael:
    And just as follow-up, can you remind us what the revenue or EBITDA expectation is for Zenium in the guide? And, I guess, based on what you just said, it doesn't sound like there's a need for equity in the second half and we should consider that more opportunistic?
  • Diane M. Morefield:
    Near-term, we don't have any need for equity again. We're just bidding on liquidity. I don't have the exact number.
  • Michael Schafer:
    Yeah. I mean I'll be...
  • Diane M. Morefield:
    Michael can follow up.
  • Michael Schafer:
    We announced the acquisition. We had indicated that we expected full-year Zenium would generate EBITDA of $17 million to $19 million. And so, you can just do the math on that for what you – the expected contribution will be for the last three months of the year.
  • Colby Synesael:
    Great. Thank you.
  • Gary J. Wojtaszek:
    Sure.
  • Operator:
    Our next question comes from Amir Rozwadowski with Barclays. Please go ahead.
  • Amir Rozwadowski:
    Yeah. Hi. Good morning, folks.
  • Gary J. Wojtaszek:
    Good morning, Roz.
  • Amir Rozwadowski:
    A couple of quick questions, if I may. Good morning. A couple of quick questions, if I may. Just on broader strategy, Gary, if we go back a couple of years, there was a lot of concerns in terms of how you folks in the broader industry in and of itself would play out with respect to the cloud opportunity. Based on some of the comments that you guys are making today, we're seeing longer term deals signed, global deals signed, in terms of expanding capacity going forward. Do you consider that sort of phase of the industry's evolution kind of gone at this point in terms of sort of your positioning? And then, probably broader from a global positioning perspective for you guys, clearly looking forward to the close of Zenium, you obviously have the partnership on the GDS side. Do you consider your footprint sort of adequate from a global perspective at this point to really tap into sort of that next evolution of deals that will come through?
  • Gary J. Wojtaszek:
    Sure. Yeah. So, look I'm a big believer in money always goes where it's most wanted. And I think what you're seeing here by the cloud companies is just a recognition that the returns that they get on investing and creating new disruptive businesses whether it's cloud or AI or driverless cars or all the other opportunities out there. They make substantially more money creating those new opportunities for themselves than tying up a lot of capital and really expensive real estate that to be honest it doesn't give them any advantage and we can deliver it faster and more inexpensively than they can repeatedly wherever they want. And so, I think what you're seeing is just the evolution of their comfort and recognition that from a capital asset perspective, they make a much higher return investing in the other parts of their business than in data centers. And I have been consistent in that view for five years. I have never backed off of that. I've always felt that the cloud companies are always going to be really strong customers of ours as they were going to outsource more and more over time and I think that's what you're seeing in the industry broadly here. With regard to our strategic ambitions, look, I have also been consistent in that I believe that this industry is global in nature. I believe that these are the largest companies in the world. And then if you really want to be as helpful as possible to those companies, you have to be global in all advanced supply chains and companies that you think world-class supply chains. The name of the game there is to reduce my supplier footprint, right? It is not efficient for cloud companies to deal with 60 or 70 different vendors around the world. They should be dealing with three vendors. And our objective is to basically be that third player in the space, and we expect that as we're catching what's going on in the cloud wave, we're really, really well-positioned because our product and our capabilities, I think, are unique and I think what you're seeing that in our success domestically is just evidence of that. And I think once we take that party on the road, you're going to see us do the same thing in Europe. And as I mentioned, we've got 250 megawatts of capacity that we'll be bringing on line in Europe. We will be a very substantial player there in no time and we expect that after, that you should expect us to do more things in Asia and the rest of the world. [Foreign Language] (49
  • Amir Rozwadowski:
    Great. Thanks very much for the incremental color. Excellent. Thanks a lot.
  • Operator:
    Our next question comes from Eric Luebchow with Wells Fargo. Please go ahead.
  • Eric Luebchow:
    Hi. Thanks for taking the question. Just curious how you feel after the – about your current level of inventory, given your really strong leasing year-to-date. And in fact, you don't have a lot of currently available supply right now in some key markets like Northern Virginia, Phoenix, and San Antonio Do you think have enough inventory on hand and under development to kind of sustain elevated leasing levels and what do you think the right amount is so that you have plenty of capacity to pursue some of these larger footprint deals? Thanks.
  • Gary J. Wojtaszek:
    Sure. Yeah. Thanks, Eric. The sales success is really – we work with a phenomenal team in the sales organization. So, the kudos goes out to Tesh and Gould and the rest of the guys there. So, with regard to inventory, I mean, we got 500,000 square feet of inventory coming on line right now. So, we've got, I think, adequate amount of inventory available to meet that demand. What you just saw last week we closed an additional property in Northern Virginia. A couple of weeks before that, we closed another property out in Arizona. So, I think we're positioned really well in this country and, for that, we're always looking for additional capacity in the key markets. We've been looking at Santa Clara for a long time. That's the only market in the U.S. where I think that we are not serving our customers as well as we could be. So, we're always looking for inventory available in that market. But outside of that, really, no other plans to do any future expansions outside of the existing markets in the U.S. where we currently are. Our whole focus is getting to Europe as fast as possible at scale so we can meet that demand because we want to position ourselves well for really nice growth into 2020 and beyond.
  • Diane M. Morefield:
    Yeah. And we're sitting on almost 400 acres of land inventory for future development. In addition, when we gave guidance at the beginning of the year, we highlighted that we had budgeted in $170 million range for land. So, we'll have future inventory as we build out on those various sites across the country and internationally.
  • Eric Luebchow:
    Got it. Thank you.
  • Operator:
    Our next question comes from Ari Klein with BMO Capital Markets. Please go ahead.
  • Aryeh Klein:
    Thanks. Going to back to leasing, what would the pipeline look like on a year-over-year basis excluding the 25 Chinese deal since I don't think that was there a year ago pre-GDS? And then, it also – it looks like about half the backlog is related to Microsoft. Can you talk about the visibility into future projects with them, and do you expect them to be in digest mode for the next little bit?
  • Gary J. Wojtaszek:
    Yeah. I mean, the funnel for the Chinese companies is it's up a little bit versus -- in aggregate versus where the funnel was last year. But it's still not meaningful in terms of the dollar amounts in there. I mean, most of that 25 are enterprise customers. So, those are typically below 500 kilowatts in deployment. So, it's not as big relatively speaking. But in aggregate, I mean, the funnel is up 23% year-over-year. I mean if you took off maybe 10% increase on the funnel associated with the Chinese deal that we're searching is probably a good estimate. And what was the second part of your question?
  • Aryeh Klein:
    Microsoft. It looks like it's about half of your backlog. So, can you just talk about into your visibility with future projects with them, and would you expect them to be in digest mode for the next little bit?
  • Gary J. Wojtaszek:
    Yeah. I think -- but we don't really specifically talk about customers per se but I would say that that is a record backlog for us. We do expect that most of that is going to get rolled out this year. And by the first quarter of next year, we feel really good about what that's going to do for our business heading into 2019 and 2020. We still have two quarters left of selling so we feel that we're positioned really well for the future. I think with regard to weather like this is digesting mode or not, I think – everyone thinks about this and I just don't think – I don't think there's a broad connection yet made between the success of the cloud companies and what the derivative impact is to the data center providers that support them. So this quarter, what you saw is AWS grew at 49%. They are a $25 billion company and that was the third sequential quarterly increase in their growth year-over-year and their $25 billion is drawing at 49%. Google put up similar results. Amazon put up similar results. Apple did. Microsoft did. Everyone is putting up phenomenal results and everyone seems to be pretty comfortable with the underlying growth profile for those companies. And yet, there just seems to be this disconnect between what's going on in the data center space relative to the cloud space. Because when you think about what we simply do, all we are doing is building these digital factories that help these cloud companies grow and we're in the business of building digital factories and the factories that we are building and selling are the same factories that are powering all the growth in all of those cloud companies. So as we sit here today, I expect that the cloud adoption is going to continue to accelerate. We're still in the early stages of the enterprise adoption. That's going to significantly accelerate over the next couple of years. And I think the bigger story here that people still haven't fully grasp is just how big the market for artificial intelligence is going to be. I think that's going to be an addressable market that's going to be three times larger than the cloud is. So, I don't see this slowing down. I think what you will see quarter to quarter is some lumpiness in sales, but in aggregate, it's going up. I think what you saw broadly is that the industry did 50% more sales this quarter than we did last quarter and that was pretty much across the board.
  • Aryeh Klein:
    Great. Thank you.
  • Operator:
    Our next question comes from Erik Rasmussen with Stifel. Please go ahead.
  • Erik Peter Rasmussen:
    Yeah. Thanks and congrats on the leasing – record leasing numbers. So, in terms of the Chinese hyperscales leasing, certainly it was a nice addition this quarter. Is that indicative of the size of the deals that we could expect to see in the future? And then with that, was it from the top – any of the top three hyperscalers or are you able to say where it came from? And then I have a follow-up.
  • Gary J. Wojtaszek:
    Yeah. Look, I mean, GDS just dominates the Chinese market, right? They have strong relationships with all the major cloud companies throughout China. They have been noted by Tencent and Alibaba as a critical supplier winning awards. So, they have phenomenal relationships and that was one of the investment thesis considerations that we did when we went into that arrangement with GDS. It's been a fantastic partnership with them and I think what you're seeing in the success that we're having in our cloud companies -- the Chinese cloud companies, that's a direct indication of how successful and strong relationship we have with GDS. I expect that what you're going to see with the Chinese cloud companies in the U.S. is going to accelerate. They're still in the early stages here relative to their U.S. competitors. But I expect you're going to see them continue to grow here in Europe and all around the world. I mean none of the CEOs that are running any of these cloud companies worldwide have small ambitions. They all want to do big things for their companies and our job is to help all those companies be successful.
  • Erik Peter Rasmussen:
    Thanks. And then just real quick getting back to kind of the some of the inventory discussion, we all saw H5 make the acquisition of Intuit's data center in Quincy. Does that kind of change your conversations with potential customers there and the potential timeline for any development there?
  • Gary J. Wojtaszek:
    No. No. We compete against everyone all over and so it's just another player in a market that we'll have to compete against.
  • Operator:
    Our next question comes from Jon Petersen with Jefferies. Please go ahead.
  • Jonathan M. Petersen:
    Great. Thanks. So I was looking at your top tenant list number six and number eight, a telecom services company and energy company. Both have about 2% of revenue. Both have about less than four months left on their leases. Curious how lease negotiations are going there or I know Diane mentioned in her prepared remarks that churn was going to be higher in the second half. Are those the two customers that we're looking at that's going to drive that higher churn?
  • Gary J. Wojtaszek:
    Yes. Yeah. One of those customers that is absolutely in our forecast for churn for the year. That company have been working on a three-year initiative to move their footprint to the cloud. The other one we will renew just at a lower rate.
  • Jonathan M. Petersen:
    Okay. I guess so what are the others? You guys have done about 2% churn in the first part of the year. It looks like another 2% will come from one of these guys. What are kind of the other drivers that get you to that 68% for the full-year?
  • Diane M. Morefield:
    Well, it's across the board. And remember, we state churn in a very conservative manner. It's gross churn. So, around half of it is pure lease terminations, move-outs, et cetera, and about half is rate adjustment. But again, the rate adjustments often are because they're taking a bigger footprint or they're not only renewing that lease but signing the lease at another data center. So, our churn is anything that affects revenue in a downward fashion. So, that's why when we say the 68%, it's not that we're losing 68% in terms of leases not renewing, it's just total churn in – on a gross dollar value.
  • Gary J. Wojtaszek:
    Yeah. It's incredibly rare for a company to actually ever leave, right? Our churn measure is really reflective of customers growing with us and getting a commensurate price reduction.
  • Jonathan M. Petersen:
    Okay. All right. That's helpful color. And then just one more on the $65 million of revenue that you guys signed this quarter. I guess what's – can you give us maybe a little more color on what's the total fully built-out cost there, I guess, including the land to – that will be required – capital be required to actually deliver that amount of revenue?
  • Gary J. Wojtaszek:
    Yeah. I mean, if you assume roughly $250 million, $5 million mag on that, but rough order of magnitude, I guess.
  • Jonathan M. Petersen:
    And then, just to clarify because I feel like I have to ask this question every once in a while. The low-teens development yield you talk about, is that on an all-in basis including the land?
  • Diane M. Morefield:
    Yes.
  • Gary J. Wojtaszek:
    Yes.
  • Diane M. Morefield:
    Land...
  • Gary J. Wojtaszek:
    Everything...
  • Diane M. Morefield:
    ...fully loaded. Capitalized infras, leasing commissions, it's a fully-loaded basis.
  • Gary J. Wojtaszek:
    Yeah. We have a video on our website which we shared a couple years ago about how we're able to achieve such low-build cost. And for those of you who may be new to the story, you guys should take a look at that because we're very transparent in terms of explaining everything we do, what's in it, what's not to do and how the magic actually happens.
  • Jonathan M. Petersen:
    Great. Thank you for the color. Appreciate it.
  • Operator:
    Our next question comes from Nick Del Deo with MoffettNathanson. Please go ahead.
  • Nicholas Ralph Del Deo:
    Hi. Thanks for fitting me in. You noted your interconnection bookings were driven by ecosystem development and SDN adoptions. I wanted to drill into that a little bit. Now, when we think about the underlying A to B side pairings, how did your mix of bookings split out between enterprise to traditional carrier versus enterprise to SDN provider versus other pairings and how has that mix changed over the past couple of years?
  • Gary J. Wojtaszek:
    Our mix really has never changed. I think what you're seeing here is just the broad trends that are happening in the industry. And that as cloud becomes a much more significant factor in the industry, as more compute nodes are deployed in our facilities, the network access point which are typically outside of our facilities is being driven back into our facility. So, I think what you're seeing is a migration of the network topology changing pretty quickly for us as the network nodes are now being deployed in our facilities. And so, what more and more of our cloud companies are recognizing is that the success that we've been having in generating and attracting enterprise companies is also the same exact customers that all the cloud companies want to sell to. And so, commercially they are seeing the opportunities to partner with us to attack that. Our enterprise companies are seeing the opportunity for them to help figure out how much of their IT stack is going to be moving to the cloud by making it easy for them to connect in our facilities. And I expect over time as the cloud more so becomes a bigger part of the IT stack, you're going to see a significant change in the network topology in terms of how people are accessing it and how those cloud companies are wanting to sell it and distribute their product to their customers.
  • Nicholas Ralph Del Deo:
    Okay. That's helpful. And then one on Zenium. Just the delay and having the stop date so close introduce any risk that another buyer might try to step in with a better offer or Zenium might try to ask for a higher price given its performance or is that a non-issue?
  • Gary J. Wojtaszek:
    Look, we've got a great relationship with the Soros guys. I mean, we've been really fair partners. We've been working really intently. Both of our intentions here are to get the deal done as quickly as possible. This delay in this process is not good. It's painful. It's really slowing down in terms of what we had wanted to do with those assets in our hands. So, I've heard that same chatter and to me this is just the same typical back story gossip that you hear on so many other things.
  • Diane M. Morefield:
    But to be clear, any new buyer of any assets like that would have to start from scratch with the regulatory process. And we've been in the process since January. So, nobody is going to be able to do it quicker. They would be starting from scratch.
  • Nicholas Ralph Del Deo:
    Okay. Well, thank you both.
  • Gary J. Wojtaszek:
    Sure.
  • Operator:
    Our next question comes from Jonathan Atkin with RBC. Please go ahead.
  • Jonathan Atkin:
    Thanks. So, I was interested in any kind of update around joint ventures. You've talked about it on some prior calls. And anything that you can provide in the way of updates on talks that maybe underway around financing based on existing cash flows or prospective developments.
  • Diane M. Morefield:
    I think it's fair to say, Jonathan, we're always looking at are there other baskets of capital. And clearly there's a ton of money wanting to get into data center space. So I think it's fair to say that we are proactive in talking to various potential capital partners that over time might come to fruition. But there's nothing really to specifically discuss at this point.
  • Gary J. Wojtaszek:
    I think in general though like if you have a lot of capital deployed, you would want to partner up with someone who knows how to deliver the lowest cost facilities in the country at the fastest time to market. I think we can sell it out faster than anyone else. So those would be the qualities that I'd look for in an operator and I think that we offer that to most of the financial players out there.
  • Jonathan Atkin:
    And then internationally looking beyond Europe, maybe to Asia, any kind of further thoughts on entering that market from getting assets there or potentially increasing your existing financial exposure in China. Thanks.
  • Gary J. Wojtaszek:
    Yeah. I mean we are absolutely looking at expanding in China – I mean not in China but in Asia. Our focus right now is Europe. We need to get a much larger platform there and a sizable presence. That's our priority. But we continue to talk to different folks in Asia as we look what the next steps of our expansion plans are going to be. As I've said, historically, I mean the original investment that we made in GDS was really done because we thought that it was a great opportunity to partner with a fantastic company there. That's gone extremely well. We feel really, really good about the success we've been able to generate with them. They would be an ideal partner with us to go jointly together throughout Asia as they could bring Chinese customers into these facilities, and we could bring our U.S. customers. So, that would be the ideal partner for us to work together throughout Asia. But for the time being, our priority is really focused on expanding in Europe.
  • Jonathan Atkin:
    Okay. And then on the interconnect question, just a kind of a follow-up. I think in broad strokes you mentioned that the drivers are very similar, but you do have companies like Megaport and PacketFabric in some of your sites. And is there any way to sort of ballpark what portion of the interconnect contribution relates to their presence inside your company? Thanks.
  • Gary J. Wojtaszek:
    Yeah. I do know that number. I don't know. I didn't look at that before the call. What I can tell you is that the Megaport growth year-over-year is enormous. Like we are seeing really big upticks from their customers and our facilities doing that. That product that they offer is very disruptive, right? They are enabling a lot of the enterprise customers to get access to various cloud vendors in a really slick easy-to-use format and by the drink. And so, that resonates really well with a lot of the enterprise companies. And we've been very good partners with Megaport. I think they're deployed in about close to 15 of our facilities right now, and the growth in aggregate with them is high. I don't know across the board percentage-wise how much of it is Megaport.
  • Jonathan Atkin:
    Okay. And then, lastly, the quarter-on-quarter trend in base revenue, the $3 million delta, does that include in the impact of, I think, the Marsh property and the other sort of non-core assets that Diane referred to or is there something else going on in terms of the existing base of business that would have led to that $3 million change?
  • Gary J. Wojtaszek:
    Are you talking about the sequential decline, Jon?
  • Jonathan Atkin:
    Yes.
  • Gary J. Wojtaszek:
    Yes. So a couple of things. One is Diane mentioned in her prepared remarks we had lease termination fees of $5 million in the first quarter. And then the impact of the two facilities that we exited, as Diane mentioned in her prepared remarks as well, that would have been captured in our churn metric.
  • Jonathan Atkin:
    Okay. Thank you.
  • Gary J. Wojtaszek:
    Yes.
  • Operator:
    Our next question comes from Jordan Sadler with KeyBanc Capital Markets. Please go ahead.
  • Jordan Sadler:
    Thanks for hanging in there.
  • Diane M. Morefield:
    Hi, Jordan.
  • Jordan Sadler:
    Hi. Can you guys parse the volume a little bit more for me? I think you guys mentioned four cloud customers in the quarter. Were there a couple of larger leases that you could flag out of this big volume?
  • Gary J. Wojtaszek:
    Yeah. I mean all those were pretty big deals with those quarters. I mean in the aggregate, was it like 90% or 80% of our bookings this quarter was...
  • Michael Schafer:
    86% was cloud.
  • Gary J. Wojtaszek:
    86% was cloud.
  • Jordan Sadler:
    So, 86% came from those four potentially?
  • Diane M. Morefield:
    Well, no. Just broadly from cloud. But yeah, obviously, the four were what weighted the most.
  • Jordan Sadler:
    Okay. You mentioned that two of the Chinese customers took about 10 megawatts. Were there any leases bigger than those or were those sort of the largest?
  • Gary J. Wojtaszek:
    No. Those were the largest.
  • Jordan Sadler:
    Okay. Do you see any leasing within the Zenium portfolio in the quarter?
  • Gary J. Wojtaszek:
    Minimal. Minimal amount of leasing.
  • Jordan Sadler:
    Okay. And then just on the 13% yield, just coming back to the pricing, is that an initial, Diane, or is that a stabilized yield? The initial cash or is it...
  • Diane M. Morefield:
    It's a stabilized development yield unlevered.
  • Jordan Sadler:
    Okay. Okay. And so, with these leases, how should we think about the average? I see that the term went up a ton. What's – what are the bumps look like or the escalators look like in these leases?
  • Diane M. Morefield:
    We get 1% to 3% in escalators particularly on the longer term leases.
  • Jordan Sadler:
    1% to 3%, okay. And then lastly, leverage. Did you say what the pro forma leverage is for Zenium once you close that you're 4.7 net debt to EBITDA now?
  • Diane M. Morefield:
    No, we haven't stated that but...
  • Gary J. Wojtaszek:
    But we're obviously well below...
  • Diane M. Morefield:
    Yeah. We're starting at 4.7 and will draw on the term loan and the line.
  • Jordan Sadler:
    Okay. And that's a $450 million-ish number, right, to close?
  • Diane M. Morefield:
    I'm sorry, Jordan. I missed that.
  • Jordan Sadler:
    The total proceeds for that are a little over $450 million?
  • Gary J. Wojtaszek:
    Yeah.
  • Diane M. Morefield:
    Yeah. It's roughly in the $0.5 billion range.
  • Jordan Sadler:
    Okay. Thank you.
  • Gary J. Wojtaszek:
    Thanks.
  • Diane M. Morefield:
    Welcome.
  • Operator:
    And this concludes our question-and-answer session. I'd like to turn the conference back over to Gary Wojtaszek for any closing remarks.
  • Gary J. Wojtaszek:
    Thanks, everyone. Appreciate you making time to talk with us this morning. We really appreciate your support and to all of CyrusOne team, this was just a phenomenal quarter. We feel really strong about us. So, thank you all very much. Thanks, everyone. Have a good weekend.
  • Operator:
    The conference has now concluded. Thank you for attending today's presentation. You may now disconnect.